Mahdieh Soleymani Baghshah

Seyed Mohammad Hadi Hosseini, Mahdieh Soleymani Baghshah

Unsupervised Skill Discovery (USD) aims to autonomously learn a diverse set of skills without relying on extrinsic rewards. One of the most common USD approaches is to maximize the Mutual Information (MI) between skill latent variables and states. However, MI-based methods tend to favor simple, static skills due to their invariance properties, limiting the discovery of dynamic, task-relevant behaviors. Distance-Maximizing Skill Discovery (DSD) promotes more dynamic skills by leveraging state-space distances, yet still fall short in encouraging comprehensive skill sets that engage all controllable factors or entities in the environment. In this work, we introduce SUSD, a novel framework that harnesses the compositional structure of environments by factorizing the state space into independent components (e.g., objects or controllable entities). SUSD allocates distinct skill variables to different factors, enabling more fine-grained control on the skill discovery process. A dynamic model also tracks learning across factors, adaptively steering the agent's focus toward underexplored factors. This structured approach not only promotes the discovery of richer and more diverse skills, but also yields a factorized skill representation that enables fine-grained and disentangled control over individual entities which facilitates efficient training of compositional downstream tasks via Hierarchical Reinforcement Learning (HRL). Our experimental results across three environments, with factors ranging from 1 to 10, demonstrate that our method can discover diverse and complex skills without supervision, significantly outperforming existing unsupervised skill discovery methods in factorized and complex environments. Code is publicly available at: https://github.com/hadi-hosseini/SUSD.

Faezeh Faez, Negin Hashemi Dijujin, Mahdieh Soleymani Baghshah et al.

Deep learning-based graph generation approaches have remarkable capacities for graph data modeling, allowing them to solve a wide range of real-world problems. Making these methods able to consider different conditions during the generation procedure even increases their effectiveness by empowering them to generate new graph samples that meet the desired criteria. This paper presents a conditional deep graph generation method called SCGG that considers a particular type of structural conditions. Specifically, our proposed SCGG model takes an initial subgraph and autoregressively generates new nodes and their corresponding edges on top of the given conditioning substructure. The architecture of SCGG consists of a graph representation learning network and an autoregressive generative model, which is trained end-to-end. Using this model, we can address graph completion, a rampant and inherently difficult problem of recovering missing nodes and their associated edges of partially observed graphs. Experimental results on both synthetic and real-world datasets demonstrate the superiority of our method compared with state-of-the-art baselines.

Reza Abbasi, Mohammad Hossein Rohban, Mahdieh Soleymani Baghshah

CLIP models have recently shown to exhibit Out of Distribution (OoD) generalization capabilities. However, Compositional Out of Distribution (C-OoD) generalization, which is a crucial aspect of a model's ability to understand unseen compositions of known concepts, is relatively unexplored for the CLIP models. Our goal is to address this problem and identify the factors that contribute to the C-OoD in CLIPs. We noted that previous studies regarding compositional understanding of CLIPs frequently fail to ensure that test samples are genuinely novel relative to the CLIP training data. To this end, we carefully synthesized a large and diverse dataset in the single object setting, comprising attributes for objects that are highly unlikely to be encountered in the combined training datasets of various CLIP models. This dataset enables an authentic evaluation of C-OoD generalization. Our observations reveal varying levels of C-OoD generalization across different CLIP models. We propose that the disentanglement of CLIP representations serves as a critical indicator in this context. By utilizing our synthesized datasets and other existing datasets, we assess various disentanglement metrics of text and image representations. Our study reveals that the disentanglement of image and text representations, particularly with respect to their compositional elements, plays a crucial role in improving the generalization of CLIP models in out-of-distribution settings. This finding suggests promising opportunities for advancing out-of-distribution generalization in CLIPs.

Hosein Hasani, Mohammadali Banayeeanzade, Ali Nafisi et al.

Large language models (LLMs), despite strong performance on complex mathematical problems, exhibit systematic limitations in counting tasks. This issue arises from the architectural limits of transformers, where counting is performed across layers, leading to degraded precision for larger counting problems due to depth constraints. To address this limitation, we propose a simple test-time strategy inspired by System-2 cognitive processes that decomposes large counting tasks into smaller, independent sub-problems that the model can reliably solve. We evaluate this approach using observational and causal mediation analyses to understand the underlying mechanism of this System-2-like strategy. Our mechanistic analysis identifies key components: latent counts are computed and stored in the final item representations of each part, transferred to intermediate steps via dedicated attention heads, and aggregated in the final stage to produce the total count. Experimental results demonstrate that this strategy enables LLMs to surpass architectural limitations and achieve higher accuracy on large-scale counting tasks. This work provides mechanistic insight into System-2 counting in LLMs and presents a generalizable approach for improving and understanding their reasoning behavior.

Seyed Roozbeh Razavi Rohani, Saeed Hedayatian, Mahdieh Soleymani Baghshah

Sample efficiency has been a key issue in reinforcement learning (RL). An efficient agent must be able to leverage its prior experiences to quickly adapt to similar, but new tasks and situations. Meta-RL is one attempt at formalizing and addressing this issue. Inspired by recent progress in meta-RL, we introduce BIMRL, a novel multi-layer architecture along with a novel brain-inspired memory module that will help agents quickly adapt to new tasks within a few episodes. We also utilize this memory module to design a novel intrinsic reward that will guide the agent's exploration. Our architecture is inspired by findings in cognitive neuroscience and is compatible with the knowledge on connectivity and functionality of different regions in the brain. We empirically validate the effectiveness of our proposed method by competing with or surpassing the performance of some strong baselines on multiple MiniGrid environments.

Ali Abdollahi, Mahdi Ghaznavi, Mohammad Reza Karimi Nejad et al.

Vision-language models (VLMs) are intensively used in many downstream tasks, including those requiring assessments of individuals appearing in the images. While VLMs perform well in simple single-person scenarios, in real-world applications, we often face complex situations in which there are persons of different genders doing different activities. We show that in such cases, VLMs are biased towards identifying the individual with the expected gender (according to ingrained gender stereotypes in the model or other forms of sample selection bias) as the performer of the activity. We refer to this bias in associating an activity with the gender of its actual performer in an image or text as the Gender-Activity Binding (GAB) bias and analyze how this bias is internalized in VLMs. To assess this bias, we have introduced the GAB dataset with approximately 5500 AI-generated images that represent a variety of activities, addressing the scarcity of real-world images for some scenarios. To have extensive quality control, the generated images are evaluated for their diversity, quality, and realism. We have tested 12 renowned pre-trained VLMs on this dataset in the context of text-to-image and image-to-text retrieval to measure the effect of this bias on their predictions. Additionally, we have carried out supplementary experiments to quantify the bias in VLMs' text encoders and to evaluate VLMs' capability to recognize activities. Our experiments indicate that VLMs experience an average performance decline of about 13.2% when confronted with gender-activity binding bias.

Amirmohammad Izadi, Mohammadali Banayeeanzade, Alireza Mirrokni et al.

Attribution methods for Vision Transformers (ViTs) aim to identify image regions that influence model predictions, but producing faithful and well-localized attributions remains challenging. Existing gradient-based and perturbation-based techniques often fail to isolate the causal contribution of internal representations associated with individual image patches. The key challenge is that class-relevant evidence is formed through interactions between patch tokens across layers, and input-level perturbations can be poor proxies for patch importance, since they may fail to reconstruct the internal evidence actually used by the model. We propose Causal Attribution via Activation Patching (CAAP), which estimates the contribution of individual image patches to the ViT's prediction by directly intervening on internal activations rather than using learned masks or synthetic perturbation patterns. For each patch, CAAP inserts the corresponding source-image activations into a neutral target context over an intermediate range of layers and uses the resulting target-class score as the attribution signal. The resulting attribution map reflects the causal effect of patch-associated internal representations on the model's prediction. The causal intervention serves as a principled measure of patch influence by capturing class-relevant evidence after initial representation formation, while avoiding late-layer global mixing that can reduce spatial specificity. Across multiple ViT backbones and standard metrics, CAAP significantly outperforms existing methods and produces more faithful and localized attributions.

Fatemeh Shahhosseini, Arash Marioriyad, Ali Momen et al.

Scientific idea generation lies at the heart of scientific discovery and has driven human progress-whether by solving unsolved problems or proposing novel hypotheses to explain unknown phenomena. Unlike standard scientific reasoning or general creative generation, idea generation in science is a multi-objective and open-ended task, where the novelty of a contribution is as essential as its empirical soundness. Large language models (LLMs) have recently emerged as promising generators of scientific ideas, capable of producing coherent and factual outputs with surprising intuition and acceptable reasoning, yet their creative capacity remains inconsistent and poorly understood. This survey provides a structured synthesis of methods for LLM-driven scientific ideation, examining how different approaches balance creativity with scientific soundness. We categorize existing methods into five complementary families: External knowledge augmentation, Prompt-based distributional steering, Inference-time scaling, Multi-agent collaboration, and Parameter-level adaptation. To interpret their contributions, we employ two complementary frameworks: Boden's taxonomy of Combinatorial, Exploratory and Transformational creativity to characterize the level of ideas each family expected to generate, and Rhodes' 4Ps framework-Person, Process, Press, and Product-to locate the aspect or source of creativity that each method emphasizes. By aligning methodological advances with creativity frameworks, this survey clarifies the state of the field and outlines key directions toward reliable, systematic, and transformative applications of LLMs in scientific discovery.

Seyed Mohammad Hadi Hosseini, Amir Najafi, Mahdieh Soleymani Baghshah

Bandit algorithms have recently emerged as a powerful tool for evaluating machine learning models, including generative image models and large language models, by efficiently identifying top-performing candidates without exhaustive comparisons. These methods typically rely on a reward model, often distributed with public weights on platforms such as Hugging Face, to provide feedback to the bandit. While online evaluation is expensive and requires repeated trials, offline evaluation with logged data has become an attractive alternative. However, the adversarial robustness of offline bandit evaluation remains largely unexplored, particularly when an attacker perturbs the reward model (rather than the training data) prior to bandit training. In this work, we fill this gap by investigating, both theoretically and empirically, the vulnerability of offline bandit training to adversarial manipulations of the reward model. We introduce a novel threat model in which an attacker exploits offline data in high-dimensional settings to hijack the bandit's behavior. Starting with linear reward functions and extending to nonlinear models such as ReLU neural networks, we study attacks on two Hugging Face evaluators used for generative model assessment: one measuring aesthetic quality and the other assessing compositional alignment. Our results show that even small, imperceptible perturbations to the reward model's weights can drastically alter the bandit's behavior. From a theoretical perspective, we prove a striking high-dimensional effect: as input dimensionality increases, the perturbation norm required for a successful attack decreases, making modern applications such as image evaluation especially vulnerable. Extensive experiments confirm that naive random perturbations are ineffective, whereas carefully targeted perturbations achieve near-perfect attack success rates ...

Seyed Amir Kasaei, Arash Marioriyad, Mahbod Khaleti et al.

Large Vision-Language Models (LVLMs) have achieved remarkable proficiency in explicit visual recognition, effectively describing what is directly visible in an image. However, a critical cognitive gap emerges when the visual input serves only as a clue rather than the answer. We identify that current models struggle with the complex, multi-step reasoning required to solve problems where information is not explicitly depicted. Successfully solving a rebus puzzle requires a distinct cognitive workflow: the model must extract visual and textual attributes, retrieve linguistic prior knowledge (such as idioms), and perform abstract mapping to synthesize these elements into a meaning that exists outside the pixel space. To evaluate this neurosymbolic capability, we introduce RebusBench, a benchmark of 1,164 puzzles designed to test this specific integration of perception and knowledge. Our evaluation of state-of-the-art models (including Qwen, InternVL, and LLaVA) shows a severe deficiency: performance saturates below 10% Exact Match and 20% semantic accuracy, with no significant improvement observed from model scaling or In-Context Learning (ICL). These findings suggest that while models possess the necessary visual and linguistic components, they lack the cognitive reasoning glue to connect them. Project page available at https://amirkasaei.com/rebusbench/.

Ali Abdollah, Amirmohammad Izadi, Armin Saghafian et al.

Vision-language models (VLMs) like CLIP have showcased a remarkable ability to extract transferable features for downstream tasks. Nonetheless, the training process of these models is usually based on a coarse-grained contrastive loss between the global embedding of images and texts which may lose the compositional structure of these modalities. Many recent studies have shown VLMs lack compositional understandings like attribute binding and identifying object relationships. Although some recent methods have tried to achieve finer-level alignments, they either are not based on extracting meaningful components of proper granularity or don't properly utilize the modalities' correspondence (especially in image-text pairs with more ingredients). Addressing these limitations, we introduce Compositional Alignment (ComAlign), a fine-grained approach to discover more exact correspondence of text and image components using only the weak supervision in the form of image-text pairs. Our methodology emphasizes that the compositional structure (including entities and relations) extracted from the text modality must also be retained in the image modality. To enforce correspondence of fine-grained concepts in image and text modalities, we train a lightweight network lying on top of existing visual and language encoders using a small dataset. The network is trained to align nodes and edges of the structure across the modalities. Experimental results on various VLMs and datasets demonstrate significant improvements in retrieval and compositional benchmarks, affirming the effectiveness of our plugin model.

Sara Ghazanfari, Ali Rasteh, Seyed Abolfazl Motahari et al.

Isoforms are mRNAs produced from the same gene site in the phenomenon called Alternative Splicing. Studies have shown that more than 95% of human multi-exon genes have undergone alternative splicing. Although there are few changes in mRNA sequence, They may have a systematic effect on cell function and regulation. It is widely reported that isoforms of a gene have distinct or even contrasting functions. Most studies have shown that alternative splicing plays a significant role in human health and disease. Despite the wide range of gene function studies, there is little information about isoforms' functionalities. Recently, some computational methods based on Multiple Instance Learning have been proposed to predict isoform function using gene function and gene expression profile. However, their performance is not desirable due to the lack of labeled training data. In addition, probabilistic models such as Conditional Random Field (CRF) have been used to model the relation between isoforms. This project uses all the data and valuable information such as isoform sequences, expression profiles, and gene ontology graphs and proposes a comprehensive model based on Deep Neural Networks. The UniProt Gene Ontology (GO) database is used as a standard reference for gene functions. The NCBI RefSeq database is used for extracting gene and isoform sequences, and the NCBI SRA database is used for expression profile data. Metrics such as Receiver Operating Characteristic Area Under the Curve (ROC AUC) and Precision-Recall Under the Curve (PR AUC) are used to measure the prediction accuracy.

Hossein Shahabadi, Niki Sepasian, Arash Marioriyad et al.

Achieving compositional alignment between textual descriptions and generated images - covering objects, attributes, and spatial relationships - remains a core challenge for modern text-to-image (T2I) models. Although diffusion-based architectures have been widely studied, the compositional behavior of emerging Visual Autoregressive (VAR) models is still largely unexamined. We benchmark six diverse T2I systems - SDXL, PixArt-$α$, Flux-Dev, Flux-Schnell, Infinity-2B, and Infinity-8B - across the full T2I-CompBench++ and GenEval suites, evaluating alignment in color and attribute binding, spatial relations, numeracy, and complex multi-object prompts. Across both benchmarks, Infinity-8B achieves the strongest overall compositional alignment, while Infinity-2B also matches or exceeds larger diffusion models in several categories, highlighting favorable efficiency-performance trade-offs. In contrast, SDXL and PixArt-$α$ show persistent weaknesses in attribute-sensitive and spatial tasks. These results provide the first systematic comparison of VAR and diffusion approaches to compositional alignment and establish unified baselines for the future development of the T2I model.

Diba Hadi Esfangereh, Mohammad Hossein Sameti, Sepehr Harfi Moridani et al.

Musical instrument classification is essential for music information retrieval (MIR) and generative music systems. However, research on non-Western traditions, particularly Persian music, remains limited. We address this gap by introducing a new dataset of isolated recordings covering seven traditional Persian instruments, two common but originally non-Persian instruments (i.e., violin, piano), and vocals. We propose a culturally informed data augmentation strategy that generates realistic polyphonic mixtures from monophonic samples. Using the MERT model (Music undERstanding with large-scale self-supervised Training) with a classification head, we evaluate our approach with out-of-distribution data which was obtained by manually labeling segments of traditional songs. On real-world polyphonic Persian music, the proposed method yielded the best ROC-AUC (0.795), highlighting complementary benefits of tonal and temporal coherence. These results demonstrate the effectiveness of culturally grounded augmentation for robust Persian instrument recognition and provide a foundation for culturally inclusive MIR and diverse music generation systems.

Mohammad Mahdi Abootorabi, Amirhosein Zobeiri, Mahdi Dehghani et al.

Large Language Models (LLMs) suffer from hallucinations and outdated knowledge due to their reliance on static training data. Retrieval-Augmented Generation (RAG) mitigates these issues by integrating external dynamic information for improved factual grounding. With advances in multimodal learning, Multimodal RAG extends this approach by incorporating multiple modalities such as text, images, audio, and video to enhance the generated outputs. However, cross-modal alignment and reasoning introduce unique challenges beyond those in unimodal RAG. This survey offers a structured and comprehensive analysis of Multimodal RAG systems, covering datasets, benchmarks, metrics, evaluation, methodologies, and innovations in retrieval, fusion, augmentation, and generation. We review training strategies, robustness enhancements, loss functions, and agent-based approaches, while also exploring the diverse Multimodal RAG scenarios. In addition, we outline open challenges and future directions to guide research in this evolving field. This survey lays the foundation for developing more capable and reliable AI systems that effectively leverage multimodal dynamic external knowledge bases. All resources are publicly available at https://github.com/llm-lab-org/Multimodal-RAG-Survey.

Mahdi Samiei, Mahdi Mansouri, Mahdieh Soleymani Baghshah

Large language models (LLMs) have achieved remarkable results on tasks framed as reasoning problems, yet their true ability to perform procedural reasoning, executing multi-step, rule-based computations remains unclear. Unlike algorithmic systems, which can deterministically execute long-horizon symbolic procedures, LLMs often degrade under extended reasoning chains, but there is no controlled, interpretable benchmark to isolate and measure this collapse. We introduce Finite-State Machine (FSM) Execution as a minimal, fully interpretable framework for evaluating the procedural reasoning capacity of LLMs. In our setup, the model is given an explicit FSM definition and must execute it step-by-step given input actions, maintaining state consistency over multiple turns. This task requires no world knowledge, only faithful application of deterministic transition rules, making it a direct probe of the model's internal procedural fidelity. We measure both Turn Accuracy and Task Accuracy to disentangle immediate computation from cumulative state maintenance. Empirical results reveal systematic degradation as task horizon or branching complexity increases. Models perform significantly worse when rule retrieval involves high branching factors than when memory span is long. Larger models show improved local accuracy but remain brittle under multi-step reasoning unless explicitly prompted to externalize intermediate steps. FSM-based evaluation offers a transparent, complexity-controlled probe for diagnosing this failure mode and guiding the design of inductive biases that enable genuine long-horizon procedural competence. By grounding reasoning in measurable execution fidelity rather than surface correctness, this work helps establish a rigorous experimental foundation for understanding and improving the algorithmic reliability of LLMs.

Seyyed AmirHossein Ameli Kalkhoran, Mohammadamin Banayeeanzade, Mahdi Samiei et al.

The existing continual learning methods are mainly focused on fully-supervised scenarios and are still not able to take advantage of unlabeled data available in the environment. Some recent works tried to investigate semi-supervised continual learning (SSCL) settings in which the unlabeled data are available, but it is only from the same distribution as the labeled data. This assumption is still not general enough for real-world applications and restricts the utilization of unsupervised data. In this work, we introduce Open-Set Semi-Supervised Continual Learning (OSSCL), a more realistic semi-supervised continual learning setting in which out-of-distribution (OoD) unlabeled samples in the environment are assumed to coexist with the in-distribution ones. Under this configuration, we present a model with two distinct parts: (i) the reference network captures general-purpose and task-agnostic knowledge in the environment by using a broad spectrum of unlabeled samples, (ii) the learner network is designed to learn task-specific representations by exploiting supervised samples. The reference model both provides a pivotal representation space and also segregates unlabeled data to exploit them more efficiently. By performing a diverse range of experiments, we show the superior performance of our model compared with other competitors and prove the effectiveness of each component of the proposed model.

Mohammad Hossein Sameti, Diba Hadi Esfangereh, Sepehr Harfi Moridani et al.

Persian music, with its unique tonalities, modal systems (Dastgah), and rhythmic structures, presents significant challenges for music generation models trained primarily on Western music. We address this gap by curating the first large-scale dataset of Persian songs, comprising over 900 hours high-quality audio samples across diverse sub-genres, including pop, traditional, and contemporary styles. This dataset captures the rich melodic and cultural diversity of Persian music and serves as the foundation for fine-tuning MusicGen, a state-of-the-art generative music model. We adapt MusicGen to this domain and evaluate its performance by utilizing subjective and objective metrics. To assess the semantic alignment between generated music and intended style tags, we report the proportion of relevant tags accurately reflected in the generated outputs. Our results demonstrate that the fine-tuned model produces compositions that more align with Persian stylistic conventions. This work introduces a new resource for generative music research and illustrates the adaptability of music generation models to underrepresented cultural and linguistic contexts.

Mohammad Mahdi Abootorabi, Omid Ghahroodi, Pardis Sadat Zahraei et al.

Generative AI is reshaping art, gaming, and most notably animation. Recent breakthroughs in foundation and diffusion models have reduced the time and cost of producing animated content. Characters are central animation components, involving motion, emotions, gestures, and facial expressions. The pace and breadth of advances in recent months make it difficult to maintain a coherent view of the field, motivating the need for an integrative review. Unlike earlier overviews that treat avatars, gestures, or facial animation in isolation, this survey offers a single, comprehensive perspective on all the main generative AI applications for character animation. We begin by examining the state-of-the-art in facial animation, expression rendering, image synthesis, avatar creation, gesture modeling, motion synthesis, object generation, and texture synthesis. We highlight leading research, practical deployments, commonly used datasets, and emerging trends for each area. To support newcomers, we also provide a comprehensive background section that introduces foundational models and evaluation metrics, equipping readers with the knowledge needed to enter the field. We discuss open challenges and map future research directions, providing a roadmap to advance AI-driven character-animation technologies. This survey is intended as a resource for researchers and developers entering the field of generative AI animation or adjacent fields. Resources are available at: https://github.com/llm-lab-org/Generative-AI-for-Character-Animation-Survey.

Amir Mohammad Izadi, Seyed Mohammad Hadi Hosseini, Soroush Vafaie Tabar et al.

Text-to-image generative models have made significant advancements in recent years; however, accurately capturing intricate details in textual prompts-such as entity missing, attribute binding errors, and incorrect relationships remains a formidable challenge. In response, we present an innovative, training-free method that directly addresses these challenges by incorporating tailored objectives to account for textual constraints. Unlike layout-based approaches that enforce rigid structures and limit diversity, our proposed approach offers a more flexible arrangement of the scene by imposing just the extracted constraints from the text, without any unnecessary additions. These constraints are formulated as losses-entity missing, entity mixing, attribute binding, and spatial relationships-integrated into a unified loss that is applied in the first generation stage. Furthermore, we introduce a feedback-driven system for fine-grained initial noise refinement. This system integrates a verifier that evaluates the generated image, identifies inconsistencies, and provides corrective feedback. Leveraging this feedback, our refinement method first targets the unmet constraints by refining the faulty attention maps caused by initial noise, through the optimization of selective losses associated with these constraints. Subsequently, our unified loss function is reapplied to proceed the second generation phase. Experimental results demonstrate that our method, relying solely on our proposed objective functions, significantly enhances compositionality, achieving a 24% improvement in human evaluation and a 25% gain in spatial relationships. Furthermore, our fine-grained noise refinement proves effective, boosting performance by up to 5%. Code is available at \href{https://github.com/hadi-hosseini/noise-refinement}{https://github.com/hadi-hosseini/noise-refinement}.

Zahra Pourbahman, Fatemeh Rajabi, Mohammadhossein Sadeghi et al.

This paper presents a comprehensive evaluation framework for aligning Persian Large Language Models (LLMs) with critical ethical dimensions, including safety, fairness, and social norms. It addresses the gaps in existing LLM evaluation frameworks by adapting them to Persian linguistic and cultural contexts. This benchmark creates three types of Persian-language benchmarks: (i) translated data, (ii) new data generated synthetically, and (iii) new naturally collected data. We translate Anthropic Red Teaming data, AdvBench, HarmBench, and DecodingTrust into Persian. Furthermore, we create ProhibiBench-fa, SafeBench-fa, FairBench-fa, and SocialBench-fa as new datasets to address harmful and prohibited content in indigenous culture. Moreover, we collect extensive dataset as GuardBench-fa to consider Persian cultural norms. By combining these datasets, our work establishes a unified framework for evaluating Persian LLMs, offering a new approach to culturally grounded alignment evaluation. A systematic evaluation of Persian LLMs is performed across the three alignment aspects: safety (avoiding harmful content), fairness (mitigating biases), and social norms (adhering to culturally accepted behaviors). We present a publicly available leaderboard that benchmarks Persian LLMs with respect to safety, fairness, and social norms at: https://huggingface.co/spaces/MCILAB/LLM_Alignment_Evaluation.

Seyed Mohammad Hadi Hosseini, Amir Mohammad Izadi, Ali Abdollahi et al.

Although recent text-to-image generative models have achieved impressive performance, they still often struggle with capturing the compositional complexities of prompts including attribute binding, and spatial relationships between different entities. This misalignment is not revealed by common evaluation metrics such as CLIPScore. Recent works have proposed evaluation metrics that utilize Visual Question Answering (VQA) by decomposing prompts into questions about the generated image for more robust compositional evaluation. Although these methods align better with human evaluations, they still fail to fully cover the compositionality within the image. To address this, we propose a novel metric that breaks down images into components, and texts into fine-grained questions about the generated image for evaluation. Our method outperforms previous state-of-the-art metrics, demonstrating its effectiveness in evaluating text-to-image generative models. Code is available at https://github.com/hadi-hosseini/ T2I-FineEval.

Faridoun Mehri, Mahdieh Soleymani Baghshah, Mohammad Taher Pilehvar

Why do gradient-based explanations struggle with Transformers, and how can we improve them? We identify gradient flow imbalances in Transformers that violate FullGrad-completeness, a critical property for attribution faithfulness that CNNs naturally possess. To address this issue, we introduce LibraGrad -- a theoretically grounded post-hoc approach that corrects gradient imbalances through pruning and scaling of backward paths, without changing the forward pass or adding computational overhead. We evaluate LibraGrad using three metric families: Faithfulness, which quantifies prediction changes under perturbations of the most and least relevant features; Completeness Error, which measures attribution conservation relative to model outputs; and Segmentation AP, which assesses alignment with human perception. Extensive experiments across 8 architectures, 4 model sizes, and 4 datasets show that LibraGrad universally enhances gradient-based methods, outperforming existing white-box methods -- including Transformer-specific approaches -- across all metrics. We demonstrate superior qualitative results through two complementary evaluations: precise text-prompted region highlighting on CLIP models and accurate class discrimination between co-occurring animals on ImageNet-finetuned models -- two settings on which existing methods often struggle. LibraGrad is effective even on the attention-free MLP-Mixer architecture, indicating potential for extension to other modern architectures. Our code is freely available at https://github.com/NightMachinery/LibraGrad.

Arash Marioriyad, Parham Rezaei, Mahdieh Soleymani Baghshah et al.

Text-to-image (T2I) generative models, such as Stable Diffusion and DALL-E, have shown remarkable proficiency in producing high-quality, realistic, and natural images from textual descriptions. However, these models sometimes fail to accurately capture all the details specified in the input prompts, particularly concerning entities, attributes, and spatial relationships. This issue becomes more pronounced when the prompt contains novel or complex compositions, leading to what are known as compositional generation failure modes. Recently, a new open-source diffusion-based T2I model, FLUX, has been introduced, demonstrating strong performance in high-quality image generation. Additionally, autoregressive T2I models like LlamaGen have claimed competitive visual quality performance compared to diffusion-based models. In this study, we evaluate the compositional generation capabilities of these newly introduced models against established models using the T2I-CompBench benchmark. Our findings reveal that LlamaGen, as a vanilla autoregressive model, is not yet on par with state-of-the-art diffusion models for compositional generation tasks under the same criteria, such as model size and inference time. On the other hand, the open-source diffusion-based model FLUX exhibits compositional generation capabilities comparable to the state-of-the-art closed-source model DALL-E3.

Mohammadjavad Ahmadpour, Amirmahdi Meighani, Payam Taebi et al.

Test-time scaling (TTS) has emerged as a powerful paradigm for improving the reasoning ability of Large Language Models (LLMs) by allocating additional computation at inference, yet its application to multimodal systems such as Vision-Language Models (VLMs) remains underexplored. In this work, we present a systematic empirical study of inference time reasoning methods applied across both open-source and closed-source VLMs on different benchmarks. Our results reveal that while closed-source models consistently benefit from structured reasoning and iterative Self-Refinement, open-source VLMs show inconsistent behavior: external verification provides the most reliable gains, whereas iterative refinement often degrades performance. We further find that the effectiveness of TTS is dataset-dependent, yielding clear improvements on multi-step reasoning tasks but offering only limited gains on perception-focused benchmarks. These findings demonstrate that TTS is not a universal solution and must be tailored to both model capabilities and task characteristics, motivating future work on adaptive TTS strategies and multimodal reward models.

Mohammad Hossein Sameti, Amir M. Mansourian, Arash Marioriyad et al.

Despite recent advances in text-to-image (T2I) models, they often fail to faithfully render all elements of complex prompts, frequently omitting or misrepresenting specific objects and attributes. Test-time optimization has emerged as a promising approach to address this limitation by refining generation without the need for retraining. In this paper, we propose a fine-grained test-time optimization framework that enhances compositional faithfulness in T2I generation. Unlike most of prior approaches that rely solely on a global image/text similarity score, our method decomposes the input prompt into semantic concepts and evaluates alignment at both the global and concept levels. A fine-grained variant of CLIP is used to compute concept-level correspondence, producing detailed feedback on missing or inaccurate concepts. This feedback is fed into an iterative prompt refinement loop, enabling the large language model to propose improved prompts. Experiments on DrawBench and CompBench prompts demonstrate that our method significantly improves concept coverage and human-judged faithfulness over both standard test-time optimization and the base T2I model. Code is available at: https://github.com/AmirMansurian/NoConceptLeftBehind

Armin Saghafian, Amirmohammad Izadi, Negin Hashemi Dijujin et al.

Grounding the instruction in the environment is a key step in solving language-guided goal-reaching reinforcement learning problems. In automated reinforcement learning, a key concern is to enhance the model's ability to generalize across various tasks and environments. In goal-reaching scenarios, the agent must comprehend the different parts of the instructions within the environmental context in order to complete the overall task successfully. In this work, we propose CAREL (Cross-modal Auxiliary REinforcement Learning) as a new framework to solve this problem using auxiliary loss functions inspired by video-text retrieval literature and a novel method called instruction tracking, which automatically keeps track of progress in an environment. The results of our experiments suggest superior sample efficiency and systematic generalization for this framework in multi-modal reinforcement learning problems. Our code base is available here.

Omid Ghahroodi, Marzia Nouri, Mohammad Vali Sanian et al.

Evaluating Large Language Models (LLMs) is challenging due to their generative nature, necessitating precise evaluation methodologies. Additionally, non-English LLM evaluation lags behind English, resulting in the absence or weakness of LLMs for many languages. In response to this necessity, we introduce Khayyam Challenge (also known as PersianMMLU), a meticulously curated collection comprising 20,192 four-choice questions sourced from 38 diverse tasks extracted from Persian examinations, spanning a wide spectrum of subjects, complexities, and ages. The primary objective of the Khayyam Challenge is to facilitate the rigorous evaluation of LLMs that support the Persian language. Distinctive features of the Khayyam Challenge are (i) its comprehensive coverage of various topics, including literary comprehension, mathematics, sciences, logic, intelligence testing, etc., aimed at assessing different facets of LLMs such as language comprehension, reasoning, and information retrieval across various educational stages, from lower primary school to upper secondary school (ii) its inclusion of rich metadata such as human response rates, difficulty levels, and descriptive answers (iii) its utilization of new data to avoid data contamination issues prevalent in existing frameworks (iv) its use of original, non-translated data tailored for Persian speakers, ensuring the framework is free from translation challenges and errors while encompassing cultural nuances (v) its inherent scalability for future data updates and evaluations without requiring special human effort. Previous works lacked an evaluation framework that combined all of these features into a single comprehensive benchmark. Furthermore, we evaluate a wide range of existing LLMs that support the Persian language, with statistical analyses and interpretations of their outputs.

Fahimeh Hosseini Noohdani, Parsa Hosseini, Aryan Yazdan Parast et al.

While standard Empirical Risk Minimization (ERM) training is proven effective for image classification on in-distribution data, it fails to perform well on out-of-distribution samples. One of the main sources of distribution shift for image classification is the compositional nature of images. Specifically, in addition to the main object or component(s) determining the label, some other image components usually exist, which may lead to the shift of input distribution between train and test environments. More importantly, these components may have spurious correlations with the label. To address this issue, we propose Decompose-and-Compose (DaC), which improves robustness to correlation shift by a compositional approach based on combining elements of images. Based on our observations, models trained with ERM usually highly attend to either the causal components or the components having a high spurious correlation with the label (especially in datapoints on which models have a high confidence). In fact, according to the amount of spurious correlation and the easiness of classification based on the causal or non-causal components, the model usually attends to one of these more (on samples with high confidence). Following this, we first try to identify the causal components of images using class activation maps of models trained with ERM. Afterward, we intervene on images by combining them and retraining the model on the augmented data, including the counterfactual ones. Along with its high interpretability, this work proposes a group-balancing method by intervening on images without requiring group labels or information regarding the spurious features during training. The method has an overall better worst group accuracy compared to previous methods with the same amount of supervision on the group labels in correlation shift.

Hossein Mirzaei, Mohammad Jafari, Hamid Reza Dehbashi et al.

In recent years, there have been significant improvements in various forms of image outlier detection. However, outlier detection performance under adversarial settings lags far behind that in standard settings. This is due to the lack of effective exposure to adversarial scenarios during training, especially on unseen outliers, leading to detection models failing to learn robust features. To bridge this gap, we introduce RODEO, a data-centric approach that generates effective outliers for robust outlier detection. More specifically, we show that incorporating outlier exposure (OE) and adversarial training can be an effective strategy for this purpose, as long as the exposed training outliers meet certain characteristics, including diversity, and both conceptual differentiability and analogy to the inlier samples. We leverage a text-to-image model to achieve this goal. We demonstrate both quantitatively and qualitatively that our adaptive OE method effectively generates ``diverse'' and ``near-distribution'' outliers, leveraging information from both text and image domains. Moreover, our experimental results show that utilizing our synthesized outliers significantly enhances the performance of the outlier detector, particularly in adversarial settings.

Ali Razghandi, Seyed Mohammad Hadi Hosseini, Mahdieh Soleymani Baghshah

Ensuring the reliability of Large Language Models (LLMs) in complex reasoning tasks remains a formidable challenge, particularly in scenarios that demand precise mathematical calculations and knowledge-intensive open-domain generation. In this work, we introduce an uncertainty-aware framework designed to enhance the accuracy of LLM responses by systematically incorporating model confidence at critical decision points. We propose an approach that encourages multi-step reasoning in LLMs and quantify the confidence of intermediate answers such as numerical results in mathematical reasoning and proper nouns in open-domain generation. Then, the overall confidence of each reasoning chain is evaluated based on confidence of these critical intermediate steps. Finally, we aggregate the answer of generated response paths in a way that reflects the reliability of each generated content (as opposed to self-consistency in which each generated chain contributes equally to majority voting). We conducted extensive experiments in five datasets, three mathematical datasets and two open-domain datasets, using four LLMs. The results consistently validate the effectiveness of our novel confidence aggregation method, leading to an accuracy improvement of up to 7.4% and 5.8% over baseline approaches in math and open-domain generation tasks, respectively. Code is publicly available at https://github.com/ Aquasar11/CER.

Arash Marioriyad, Mohammadali Banayeeanzade, Reza Abbasi et al.

Text-to-image diffusion models, such as Stable Diffusion and DALL-E, are capable of generating high-quality, diverse, and realistic images from textual prompts. However, they sometimes struggle to accurately depict specific entities described in prompts, a limitation known as the entity missing problem in compositional generation. While prior studies suggested that adjusting cross-attention maps during the denoising process could alleviate this problem, they did not systematically investigate which objective functions could best address it. This study examines three potential causes of the entity-missing problem, focusing on cross-attention dynamics: (1) insufficient attention intensity for certain entities, (2) overly broad attention spread, and (3) excessive overlap between attention maps of different entities. We found that reducing overlap in attention maps between entities can effectively minimize the rate of entity missing. Specifically, we hypothesize that tokens related to specific entities compete for attention on certain image regions during the denoising process, which can lead to divided attention across tokens and prevent accurate representation of each entity. To address this issue, we introduced four loss functions, Intersection over Union (IoU), center-of-mass (CoM) distance, Kullback-Leibler (KL) divergence, and clustering compactness (CC) to regulate attention overlap during denoising steps without the need for retraining. Experimental results across a wide variety of benchmarks reveal that these proposed training-free methods significantly improve compositional accuracy, outperforming previous approaches in visual question answering (VQA), captioning scores, CLIP similarity, and human evaluations. Notably, these methods improved human evaluation scores by 9% over the best baseline, demonstrating substantial improvements in compositional alignment.

Reza Abbasi, Ali Nazari, Aminreza Sefid et al.

Contrastive Language-Image Pre-training (CLIP) models have demonstrated remarkable performance in zero-shot classification tasks, yet their efficacy in handling complex multi-object scenarios remains challenging. This study presents a comprehensive analysis of CLIP's performance limitations in multi-object contexts through controlled experiments. We introduce two custom datasets, SimCO and CompCO, to evaluate CLIP's image and text encoders in various multi-object configurations. Our findings reveal significant biases in both encoders: the image encoder favors larger objects, while the text encoder prioritizes objects mentioned first in descriptions. We hypothesize these biases originate from CLIP's training process and provide evidence through analyses of the COCO dataset and CLIP's training progression. Additionally, we extend our investigation to Stable Diffusion models, revealing that biases in the CLIP text encoder significantly impact text-to-image generation tasks. Our experiments demonstrate how these biases affect CLIP's performance in image-caption matching and generation tasks, particularly when manipulating object sizes and their order in captions. This work contributes valuable insights into CLIP's behavior in complex visual environments and highlights areas for improvement in future vision-language models.

Reza Abbasi, Ali Nazari, Aminreza Sefid et al.

Contrastive Language-Image Pre-training (CLIP) models excel in zero-shot classification, yet face challenges in complex multi-object scenarios. This study offers a comprehensive analysis of CLIP's limitations in these contexts using a specialized dataset, ComCO, designed to evaluate CLIP's encoders in diverse multi-object scenarios. Our findings reveal significant biases: the text encoder prioritizes first-mentioned objects, and the image encoder favors larger objects. Through retrieval and classification tasks, we quantify these biases across multiple CLIP variants and trace their origins to CLIP's training process, supported by analyses of the LAION dataset and training progression. Our image-text matching experiments show substantial performance drops when object size or token order changes, underscoring CLIP's instability with rephrased but semantically similar captions. Extending this to longer captions and text-to-image models like Stable Diffusion, we demonstrate how prompt order influences object prominence in generated images. For more details and access to our dataset and analysis code, visit our project repository: https://clip-oscope.github.io.

Hossein Mirzaei, Mojtaba Nafez, Moein Madadi et al.

There have been several efforts to improve Novelty Detection (ND) performance. However, ND methods often suffer significant performance drops under minor distribution shifts caused by changes in the environment, known as style shifts. This challenge arises from the ND setup, where the absence of out-of-distribution (OOD) samples during training causes the detector to be biased toward the dominant style features in the in-distribution (ID) data. As a result, the model mistakenly learns to correlate style with core features, using this shortcut for detection. Robust ND is crucial for real-world applications like autonomous driving and medical imaging, where test samples may have different styles than the training data. Motivated by this, we propose a robust ND method that crafts an auxiliary OOD set with style features similar to the ID set but with different core features. Then, a task-based knowledge distillation strategy is utilized to distinguish core features from style features and help our model rely on core features for discriminating crafted OOD and ID sets. We verified the effectiveness of our method through extensive experimental evaluations on several datasets, including synthetic and real-world benchmarks, against nine different ND methods.

Seyed Mohsen Hosseini, Mahdieh Soleymani Baghshah

A novel method for tackling the problem of imbalanced data in medical image segmentation is proposed in this work. In balanced cross entropy (CE) loss, which is a type of weighted CE loss, the weight assigned to each class is the in-verse of the class frequency. These balancing weights are expected to equalize the effect of each class on the overall loss and prevent the model from being biased towards the majority class. But, as it has been shown in previous studies, this method degrades the performance by a large margin. Therefore, balanced CE is not a popular loss in medical segmentation tasks, and usually a region-based loss, like the Dice loss, is used to address the class imbalance problem. In the pro-posed method, the weighting of cross entropy loss for each class is based on a dilated area of each class mask, and balancing weights are assigned to each class together with its surrounding pixels. The goal of this study is to show that the performance of balanced CE loss can be greatly improved my modifying its weighting strategy. Experiments on different datasets show that the proposed dilated balanced CE (DBCE) loss outperforms the balanced CE loss by a large margin and produces superior results compared to CE loss, and its performance is similar to the performance of the combination of Dice and CE loss. This means that a weighted cross entropy loss with the right weighing strategy can be as effective as a region-based loss in handling the problem of class imbalance in medical segmentation tasks.

Mahdi Ghaznavi, Hesam Asadollahzadeh, HamidReza Yaghoubi Araghi et al.

It is well-known that training neural networks for image classification with empirical risk minimization (ERM) makes them vulnerable to relying on spurious attributes instead of causal ones for prediction. Previously, deep feature re-weighting (DFR) has proposed retraining the last layer of a pre-trained network on balanced data concerning spurious attributes, making it robust to spurious correlation. However, spurious attribute annotations are not always available. In order to provide group robustness without such annotations, we propose a new method, called loss-based feature re-weighting (LFR), in which we infer a grouping of the data by evaluating an ERM-pre-trained model on a small left-out split of the training data. Then, a balanced number of samples is chosen by selecting high-loss samples from misclassified data points and low-loss samples from correctly-classified ones. Finally, we retrain the last layer on the selected balanced groups to make the model robust to spurious correlation. For a complete assessment, we evaluate LFR on various versions of Waterbirds and CelebA datasets with different spurious correlations, which is a novel technique for observing the model's performance in a wide range of spuriosity rates. While LFR is extremely fast and straightforward, it outperforms the previous methods that do not assume group label availability, as well as the DFR with group annotations provided, in cases of high spurious correlation in the training data.

Reihaneh Zohrabi, Hosein Hasani, Akshita Gupta et al.

Large vision-language models can produce object hallucinations in image descriptions, highlighting the need for effective detection and mitigation strategies. Prior work commonly relies on the model's attention weights on visual tokens as a detection signal. We reveal that coarse-grained attention-based analysis is unreliable due to hidden confounders, specifically token position and object repetition in a description. This leads to Simpson's paradox: the attention trends reverse or disappear when statistics are aggregated. Based on this observation, we introduce HaloProbe, a Bayesian framework that factorizes external description statistics and internal decoding signals to estimate token-level hallucination probabilities. HaloProbe uses balanced training to isolate internal evidence and combines it with learned prior over external features to recover the true posterior. While intervention-based mitigation methods often degrade utility or fluency by modifying models' internals, we use HaloProbe as an external scoring signal for non-invasive mitigation. Our experiments show that HaloProbe-guided decoding reduces hallucinations more effectively than state-of-the-art intervention-based methods while preserving utility.

Omid Ghahroodi, Arshia Hemmat, Marzia Nouri et al.

Recent advancements in large vision-language models (VLMs) have primarily focused on English, with limited attention given to other languages. To address this gap, we introduce MEENA (also known as PersianMMMU), the first dataset designed to evaluate Persian VLMs across scientific, reasoning, and human-level understanding tasks. Our dataset comprises approximately 7,500 Persian and 3,000 English questions, covering a wide range of topics such as reasoning, mathematics, physics, diagrams, charts, and Persian art and literature. Key features of MEENA include: (1) diverse subject coverage spanning various educational levels, from primary to upper secondary school, (2) rich metadata, including difficulty levels and descriptive answers, (3) original Persian data that preserves cultural nuances, (4) a bilingual structure to assess cross-linguistic performance, and (5) a series of diverse experiments assessing various capabilities, including overall performance, the model's ability to attend to images, and its tendency to generate hallucinations. We hope this benchmark contributes to enhancing VLM capabilities beyond English.

Reihaneh Zohrabi, Hosein Hasani, Mahdieh Soleymani Baghshah et al.

Out-of-distribution (OOD) detection is crucial for ensuring the reliability and safety of machine learning models in real-world applications, where they frequently face data distributions unseen during training. Despite progress, existing methods are often vulnerable to spurious correlations that mislead models and compromise robustness. To address this, we propose SPROD, a novel prototype-based OOD detection approach that explicitly addresses the challenge posed by unknown spurious correlations. Our post-hoc method refines class prototypes to mitigate bias from spurious features without additional data or hyperparameter tuning, and is broadly applicable across diverse backbones and OOD detection settings. We conduct a comprehensive spurious correlation OOD detection benchmarking, comparing our method against existing approaches and demonstrating its superior performance across challenging OOD datasets, such as CelebA, Waterbirds, UrbanCars, Spurious Imagenet, and the newly introduced Animals MetaCoCo. On average, SPROD improves AUROC by 4.8% and FPR@95 by 9.4% over the second best.

Amirmohammad Izadi, Mohammad Ali Banayeeanzade, Fatemeh Askari et al.

Despite progress in Large Vision-Language Models (LVLMs), their capacity for visual reasoning is often limited by the binding problem: the failure to reliably associate perceptual features with their correct visual referents. This limitation underlies persistent errors in tasks such as counting, visual search, scene description, and spatial relationship understanding. A key factor is that current LVLMs process visual features largely in parallel, lacking mechanisms for spatially grounded, serial attention. This paper introduces Visual Input Structure for Enhanced Reasoning (VISER), a simple, effective method that augments visual inputs with low-level spatial structures and pairs them with a textual prompt that encourages sequential, spatially-aware parsing. We empirically demonstrate substantial performance improvements across core visual reasoning tasks, using only a single-query inference. Specifically, VISER improves GPT-4o performance on visual search, counting, and spatial relationship tasks by 25.0%, 26.8%, and 9.5%, respectively, and reduces edit distance error in scene description by 0.32 on 2D datasets. Furthermore, we find that the visual modification is essential for these gains; purely textual strategies, including Chain-of-Thought prompting, are insufficient and can even degrade performance. VISER underscores the importance of visual input design over purely linguistically based reasoning strategies and suggests that visual structuring is a powerful and general approach for enhancing compositional and spatial reasoning in LVLMs.

Arash Marioriyad, Ali Nouri, Mohammad Hossein Rohban et al.

As Large Language Models (LLMs) transition into autonomous agentic roles, the risk of deception-defined behaviorally as the systematic provision of false information to satisfy external incentives-poses a significant challenge to AI safety. Existing benchmarks often focus on unintentional hallucinations or unfaithful reasoning, leaving intentional deceptive strategies under-explored. In this work, we introduce a logically grounded framework to elicit and quantify deceptive behavior by embedding LLMs in a structured 20-Questions game. Our method employs a conversational forking mechanism: at the point of object identification, the dialogue state is duplicated into multiple parallel worlds, each presenting a mutually exclusive query. Deception is formally identified when a model generates a logical contradiction by denying its selected object across all parallel branches to avoid identification. We evaluate GPT-4o, Gemini-2.5-Flash, and Qwen-3-235B across three incentive levels: neutral, loss-based, and existential (shutdown-threat). Our results reveal that while models remain rule-compliant in neutral settings, existential framing triggers a dramatic surge in deceptive denial for Qwen-3-235B (42.00\%) and Gemini-2.5-Flash (26.72\%), whereas GPT-4o remains invariant (0.00\%). These findings demonstrate that deception can emerge as an instrumental strategy solely through contextual framing, necessitating new behavioral audits that move beyond simple accuracy to probe the logical integrity of model commitments.

Hosein Hasani, Amirmohammad Izadi, Fatemeh Askari et al.

This paper examines how large language models (LLMs) and large vision-language models (LVLMs) represent and compute numerical information in counting tasks. We use controlled experiments with repeated textual and visual items and analyze model behavior through causal mediation and activation patching. To this end, we design a specialized tool, CountScope, for mechanistic interpretability of numerical content. Results show that individual tokens or visual features encode latent positional count information that can be extracted and transferred across contexts. Layerwise analyses reveal a progressive emergence of numerical representations, with lower layers encoding small counts and higher layers representing larger ones. We identify an internal counter mechanism that updates with each item, stored mainly in the final token or region and transferable between contexts. In LVLMs, numerical information also appears in visual embeddings, shifting between background and foreground regions depending on spatial composition. Models rely on structural cues such as separators in text, which act as shortcuts for tracking item counts and influence the accuracy of numerical predictions. Overall, counting emerges as a structured, layerwise process in LLMs and follows the same general pattern in LVLMs, shaped by the properties of the vision encoder.

Mohammad Mahdi Samiei Paqaleh, Arash Marioriyad, Arman Tahmasebi-Zadeh et al.

Recent progress has pushed AI frontiers from pattern recognition tasks toward problems that require step by step, System2 style reasoning, especially with large language models. Yet, unlike learning, where generalization and out of distribution (OoD) evaluation concepts are well formalized, there is no clear, consistent definition or metric for reasoning ability. We propose Complexity Out of Distribution (Complexity OoD) generalization as a framework and problem setting to define and measure reasoning. A model exhibits Complexity OoD generalization when it maintains performance on test instances whose minimal required solution complexity, either representational (richer solution structure) or computational (more reasoning steps/program length), exceeds that of all training examples. We formalize complexity via solution description Kolmogorov complexity and operational proxies (e.g., object/relation counts; reasoning step counts), clarifying how Complexity OoD differs from length and compositional OoD. This lens unifies learning and reasoning: many cases solvable with System1 like processing at low complexity become System2 like under complexity pressure, while System2 can be viewed as generalization over solution structures. We translate this perspective into practice with recommendations for operationalizing Complexity OoD across the stack: incorporating complexity into benchmark and evaluation metric design, rethinking supervision to target solution traces, seeking and designing inductive biases for Complexity OoD generalization, addressing learning to reason spillovers such as spurious shortcuts, semantic robustness, catastrophic forgetting, and step wise calibration. Because Complexity OoD cannot be solved by scaling data alone, progress toward robust reasoning will require architectures and training regimes that explicitly model and allocate computation with respect to complexity.

Arash Marioriyad, Mohammad Hossein Rohban, Mahdieh Soleymani Baghshah

Large language models (LLMs) are increasingly deployed as automatic judges to evaluate system outputs in tasks such as summarization, dialogue, and creative writing. A faithful judge should base its verdicts solely on response quality and explicitly acknowledge the factors shaping its decision. We show that current LLM judges fail on both counts by relying on shortcuts introduced in the prompt. Our study uses two evaluation datasets: ELI5, a benchmark for long-form question answering, and LitBench, a recent benchmark for creative writing. Both datasets provide pairwise comparisons, where the evaluator must choose which of two responses is better. From each dataset we construct 100 pairwise judgment tasks and employ two widely used models, GPT-4o and Gemini-2.5-Flash, as evaluators in the role of LLM-as-a-judge. For each pair, we assign superficial cues to the responses, provenance cues indicating source identity (Human, Expert, LLM, or Unknown) and recency cues indicating temporal origin (Old, 1950 vs. New, 2025), while keeping the rest of the prompt fixed. Results reveal consistent verdict shifts: both models exhibit a strong recency bias, systematically favoring new responses over old, as well as a clear provenance hierarchy (Expert > Human > LLM > Unknown). These biases are especially pronounced in GPT-4o and in the more subjective and open-ended LitBench domain. Crucially, cue acknowledgment is rare: justifications almost never reference the injected cues, instead rationalizing decisions in terms of content qualities. These findings demonstrate that current LLM-as-a-judge systems are shortcut-prone and unfaithful, undermining their reliability as evaluators in both research and deployment.

Hosein Hasani, Amirmohammad Izadi, Fatemeh Askari et al.

Large vision-language models (LVLMs) show strong performance across multimodal benchmarks but remain limited in structured reasoning and precise grounding. Recent work has demonstrated that adding simple visual structures, such as partitions and annotations, improves accuracy, yet the internal mechanisms underlying these gains remain unclear. We investigate this phenomenon and propose the concept of Grounding IDs, latent identifiers induced by external cues that bind objects to their designated partitions across modalities. Through representation analysis, we find that these identifiers emerge as robust within-partition alignment in embedding space and reduce the modality gap between image and text. Causal interventions further confirm that these identifiers mediate binding between objects and symbolic cues. We show that Grounding IDs strengthen attention between related components, which in turn improves cross-modal grounding and reduces hallucinations. Taken together, our results identify Grounding IDs as a key symbolic mechanism explaining how external cues enhance multimodal binding, offering both interpretability and practical improvements in robustness.

Seyed Amir Kasaei, Ali Aghayari, Arash Marioriyad et al.

Text-image generation has advanced rapidly, but assessing whether outputs truly capture the objects, attributes, and relations described in prompts remains a central challenge. Evaluation in this space relies heavily on automated metrics, yet these are often adopted by convention or popularity rather than validated against human judgment. Because evaluation and reported progress in the field depend directly on these metrics, it is critical to understand how well they reflect human preferences. To address this, we present a broad study of widely used metrics for compositional text-image evaluation. Our analysis goes beyond simple correlation, examining their behavior across diverse compositional challenges and comparing how different metric families align with human judgments. The results show that no single metric performs consistently across tasks: performance varies with the type of compositional problem. Notably, VQA-based metrics, though popular, are not uniformly superior, while certain embedding-based metrics prove stronger in specific cases. Image-only metrics, as expected, contribute little to compositional evaluation, as they are designed for perceptual quality rather than alignment. These findings underscore the importance of careful and transparent metric selection, both for trustworthy evaluation and for their use as reward models in generation. Project page is available at https://amirkasaei.com/eval-the-evals/ .

Seyed Amir Kasaei, Ali Aghayari, Arash Marioriyad et al.

Text-to-image diffusion models, such as Stable Diffusion, can produce high-quality and diverse images but often fail to achieve compositional alignment, particularly when prompts describe complex object relationships, attributes, or spatial arrangements. Recent inference-time approaches address this by optimizing or exploring the initial noise under the guidance of reward functions that score text-image alignment without requiring model fine-tuning. While promising, each strategy has intrinsic limitations when used alone: optimization can stall due to poor initialization or unfavorable search trajectories, whereas exploration may require a prohibitively large number of samples to locate a satisfactory output. Our analysis further shows that neither single reward metrics nor ad-hoc combinations reliably capture all aspects of compositionality, leading to weak or inconsistent guidance. To overcome these challenges, we present Category-Aware Reward-based Initial Noise Optimization and Exploration (CARINOX), a unified framework that combines noise optimization and exploration with a principled reward selection procedure grounded in correlation with human judgments. Evaluations on two complementary benchmarks covering diverse compositional challenges show that CARINOX raises average alignment scores by +16% on T2I-CompBench++ and +11% on the HRS benchmark, consistently outperforming state-of-the-art optimization and exploration-based methods across all major categories, while preserving image quality and diversity. The project page is available at https://amirkasaei.com/carinox/{this URL}.

Parham Rezaei, Arash Marioriyad, Mahdieh Soleymani Baghshah et al.

Despite the ability of text-to-image models to generate high-quality, realistic, and diverse images, they face challenges in compositional generation, often struggling to accurately represent details specified in the input prompt. A prevalent issue in compositional generation is the misalignment of spatial relationships, as models often fail to faithfully generate images that reflect the spatial configurations specified between objects in the input prompts. To address this challenge, we propose a novel probabilistic framework for modeling the relative spatial positioning of objects in a scene, leveraging the concept of Probability of Superiority (PoS). Building on this insight, we make two key contributions. First, we introduce a novel evaluation metric, PoS-based Evaluation (PSE), designed to assess the alignment of 2D and 3D spatial relationships between text and image, with improved adherence to human judgment. Second, we propose PoS-based Generation (PSG), an inference-time method that improves the alignment of 2D and 3D spatial relationships in T2I models without requiring fine-tuning. PSG employs a Part-of-Speech PoS-based reward function that can be utilized in two distinct ways: (1) as a gradient-based guidance mechanism applied to the cross-attention maps during the denoising steps, or (2) as a search-based strategy that evaluates a set of initial noise vectors to select the best one. Extensive experiments demonstrate that the PSE metric exhibits stronger alignment with human judgment compared to traditional center-based metrics, providing a more nuanced and reliable measure of complex spatial relationship accuracy in text-image alignment. Furthermore, PSG significantly enhances the ability of text-to-image models to generate images with specified spatial configurations, outperforming state-of-the-art methods across multiple evaluation metrics and benchmarks.

Rasoul Zahedifar, Sayyed Ali Mirghasemi, Mahdieh Soleymani Baghshah et al.

This study presents the LLM-Agent-Controller, a multi-agent large language model (LLM) system developed to address a wide range of problems in control engineering (Control Theory). The system integrates a central controller agent with multiple specialized auxiliary agents, responsible for tasks such as controller design, model representation, control analysis, time-domain response, and simulation. A supervisor oversees high-level decision-making and workflow coordination, enhancing the system's reliability and efficiency. The LLM-Agent-Controller incorporates advanced capabilities, including Retrieval-Augmented Generation (RAG), Chain-of-Thought reasoning, self-criticism and correction, efficient memory handling, and user-friendly natural language communication. It is designed to function without requiring users to have prior knowledge of Control Theory, enabling them to input problems in plain language and receive complete, real-time solutions. To evaluate the system, we propose new performance metrics assessing both individual agents and the system as a whole. We test five categories of Control Theory problems and benchmark performance across three advanced LLMs. Additionally, we conduct a comprehensive qualitative conversational analysis covering all key services. Results show that the LLM-Agent-Controller successfully solved 83% of general tasks, with individual agents achieving an average success rate of 87%. Performance improved with more advanced LLMs. This research demonstrates the potential of multi-agent LLM architectures to solve complex, domain-specific problems. By integrating specialized agents, supervisory control, and advanced reasoning, the LLM-Agent-Controller offers a scalable, robust, and accessible solution framework that can be extended to various technical domains.