Bowen Sun

Hang Wu, Yujun Cai, Zehao Li et al.

Understanding camera dynamics is a fundamental pillar of video spatial intelligence. However, existing multimodal models predominantly treat this task as a black-box classification, often confusing physically distinct motions by relying on superficial visual patterns rather than geometric cues. We present CamReasoner, a framework that reformulates camera movement understanding as a structured inference process to bridge the gap between perception and cinematic logic. Our approach centers on the Observation-Thinking-Answer (O-T-A) paradigm, which compels the model to decode spatio-temporal cues such as trajectories and view frustums within an explicit reasoning block. To instill this capability, we construct a Large-scale Inference Trajectory Suite comprising 18k SFT reasoning chains and 38k RL feedback samples. Notably, we are the first to employ RL for logical alignment in this domain, ensuring motion inferences are grounded in physical geometry rather than contextual guesswork. By applying Reinforcement Learning to the Observation-Think-Answer (O-T-A) reasoning paradigm, CamReasoner effectively suppresses hallucinations and achieves state-of-the-art performance across multiple benchmarks.

Bowen Sun, Yujun Cai, Ming-Hsuan Yang et al.

Video LLMs suffer from temporal inconsistency: small shifts in frame timing can flip attention and suppress relevant frames. We trace this instability to the common extension of Rotary Position Embeddings to video through multimodal RoPE. The induced inverse Fourier time kernel exhibits frame-scale ripples that multiply adjacent frames by different factors, which perturbs attention that should otherwise be governed by the raw query key inner product. We present Phase Aggregated Smoothing (PAS), a simple, training-free mechanism that applies small opposed phase offsets across heads and then aggregates their outputs. PAS preserves the per-head spectrum magnitude, while the aggregation effectively smooths the temporal kernel and reduces phase sensitivity without changing the positional encoding structure. Our analysis shows that the RoPE rotated logit can be approximated as a content dot product scaled by a time kernel; smoothing this kernel yields Lipschitz stability of attention to small temporal shifts; multi phase averaging attenuates high frequency ripples while preserving per-head spectra under Nyquist-valid sampling. Experiments on multiple video understanding benchmarks under matched token budgets show consistent improvements with negligible computational overhead. PAS provides a plug and play upgrade for robust temporal encoding in Video LLMs.

Xiaoyu Zhang, Yishan Li, Jiayin Wang et al.

The explainability of recommender systems has attracted significant attention in academia and industry. Many efforts have been made for explainable recommendations, yet evaluating the quality of the explanations remains a challenging and unresolved issue. In recent years, leveraging LLMs as evaluators presents a promising avenue in Natural Language Processing tasks (e.g., sentiment classification, information extraction), as they perform strong capabilities in instruction following and common-sense reasoning. However, evaluating recommendation explanatory texts is different from these NLG tasks, as its criteria are related to human perceptions and are usually subjective. In this paper, we investigate whether LLMs can serve as evaluators of recommendation explanations. To answer the question, we utilize real user feedback on explanations given from previous work and additionally collect third-party annotations and LLM evaluations. We design and apply a 3-level meta evaluation strategy to measure the correlation between evaluator labels and the ground truth provided by users. Our experiments reveal that LLMs, such as GPT4, can provide comparable evaluations with appropriate prompts and settings. We also provide further insights into combining human labels with the LLM evaluation process and utilizing ensembles of multiple heterogeneous LLM evaluators to enhance the accuracy and stability of evaluations. Our study verifies that utilizing LLMs as evaluators can be an accurate, reproducible and cost-effective solution for evaluating recommendation explanation texts. Our code is available at https://github.com/Xiaoyu-SZ/LLMasEvaluator.

Yiju Guo, Ganqu Cui, Lifan Yuan et al. · tencent-ai

Alignment in artificial intelligence pursues the consistency between model responses and human preferences as well as values. In practice, the multifaceted nature of human preferences inadvertently introduces what is known as the "alignment tax" -a compromise where enhancements in alignment within one objective (e.g.,harmlessness) can diminish performance in others (e.g.,helpfulness). However, existing alignment techniques are mostly unidirectional, leading to suboptimal trade-offs and poor flexibility over various objectives. To navigate this challenge, we argue the prominence of grounding LLMs with evident preferences. We introduce controllable preference optimization (CPO), which explicitly specifies preference scores for different objectives, thereby guiding the model to generate responses that meet the requirements. Our experimental analysis reveals that the aligned models can provide responses that match various preferences among the "3H" (helpfulness, honesty, harmlessness) desiderata. Furthermore, by introducing diverse data and alignment goals, we surpass baseline methods in aligning with single objectives, hence mitigating the impact of the alignment tax and achieving improvements in multi-objective alignment.

Bowen Sun, Guo Lu, Shibao Zheng

The facial sketch synthesis (FSS) model, capable of generating sketch portraits from given facial photographs, holds profound implications across multiple domains, encompassing cross-modal face recognition, entertainment, art, media, among others. However, the production of high-quality sketches remains a formidable task, primarily due to the challenges and flaws associated with three key factors: (1) the scarcity of artist-drawn data, (2) the constraints imposed by limited style types, and (3) the deficiencies of processing input information in existing models. To address these difficulties, we propose a lightweight end-to-end synthesis model that efficiently converts images to corresponding multi-stylized sketches, obviating the necessity for any supplementary inputs (\eg, 3D geometry). In this study, we overcome the issue of data insufficiency by incorporating semi-supervised learning into the training process. Additionally, we employ a feature extraction module and style embeddings to proficiently steer the generative transformer during the iterative prediction of masked image tokens, thus achieving a continuous stylized output that retains facial features accurately in sketches. The extensive experiments demonstrate that our method consistently outperforms previous algorithms across multiple benchmarks, exhibiting a discernible disparity.

Pouria Mortezaagha, Joseph Shaw, Bowen Sun et al.

Biomedical evidence synthesis relies on accurate extraction of methodological, laboratory, and outcome variables from full-text research articles, yet these variables are embedded in complex scientific PDFs that make manual abstraction time-consuming and difficult to scale. Existing document AI systems remain limited by OCR errors, long-document fragmentation, constrained throughput, and insufficient auditability for high-stakes synthesis. We present a schema-constrained AI extraction system that transforms full-text biomedical PDFs into structured, analysis-ready records by explicitly restricting model inference through typed schemas, controlled vocabularies, and evidence-gated decisions. Documents are ingested using resume-aware hashing, partitioned into caption-aware page-level chunks, and processed asynchronously under explicit concurrency controls. Chunk-level outputs are deterministically merged into study-level records using conflict-aware consolidation, set-based aggregation, and sentence-level provenance to support traceability and post-hoc audit. Evaluated on a corpus of studies on direct oral anticoagulant level measurement, the pipeline processed all documents without manual intervention, maintained stable throughput under service constraints, and exhibited strong internal consistency across document chunks. Iterative schema refinement substantially improved extraction fidelity for synthesis-critical variables, including assay classification, outcome definitions, follow-up duration, and timing of measurement. These results demonstrate that schema-constrained, provenance-aware extraction enables scalable and auditable transformation of heterogeneous scientific PDFs into structured evidence, aligning modern document AI with the transparency and reliability requirements of biomedical evidence synthesis.

Bowen Sun, Chaozhuo Li, Yaodong Yang et al.

Decompositional jailbreaks pose a critical threat to large language models (LLMs) by allowing adversaries to fragment a malicious objective into a sequence of individually benign queries that collectively reconstruct prohibited content. In real-world deployments, LLMs face a continuous, untraceable stream of fully anonymized and arbitrarily interleaved requests, infiltrated by covertly distributed adversarial queries. Under this rigorous threat model, state-of-the-art defensive strategies exhibit fundamental limitations. In the absence of trustworthy user metadata, they are incapable of tracking global historical contexts, while their deployment of generative models for real-time monitoring introduces computationally prohibitive overhead. To address this, we present TwinGate, a stateful dual-encoder defense framework. TwinGate employs Asymmetric Contrastive Learning (ACL) to cluster semantically disparate but intent-matched malicious fragments in a shared latent space, while a parallel frozen encoder suppresses false positives arising from benign topical overlap. Each request requires only a single lightweight forward pass, enabling the defense to execute in parallel with the target model's prefill phase at negligible latency overhead. To evaluate our approach and advance future research, we construct a comprehensive dataset of over 3.62 million instructions spanning 8,600 distinct malicious intents. Evaluated on this large-scale corpus under a strictly causal protocol, TwinGate achieves high malicious intent recall at a remarkably low false positive rate while remaining highly robust against adaptive attacks. Furthermore, our proposal substantially outperforms stateful and stateless baselines, delivering superior throughput and reduced latency.

Bowen Sun, Yujun Cai, Ming-Hsuan Yang et al.

Discrete diffusion language models have shown strong potential for text generation, yet standard supervised fine-tuning (SFT) misaligns with their semi-autoregressive inference: training randomly masks tokens across the entire response, while inference generates fixed-size blocks sequentially. This mismatch introduces noisy prefixes and leaky suffixes, biasing gradients away from the desired blockwise likelihood. We propose Blockwise SFT, which partitions responses into fixed-size blocks, selects one active block per step for stochastic masking, freezes all preceding tokens, and fully hides future ones. Loss is computed only over the active block, directly mirroring the blockwise decoding process. Experiments on GSM8K, MATH, and MetaMathQA show consistent gains over classical SFT under equal compute or token budgets. Block size consistency studies and ablations confirm that improvements stem from faithful training-inference alignment rather than incidental masking effects. Our results highlight the importance of matching supervision granularity to the decoding procedure in diffusion-based language models.

Maria Rafaela Gkeka, Bowen Sun, Evgenia Smirni et al.

Continuous advancements in deep learning have led to significant progress in feature detection, resulting in enhanced accuracy in tasks like Simultaneous Localization and Mapping (SLAM). Nevertheless, the vulnerability of deep neural networks to adversarial attacks remains a challenge for their reliable deployment in applications, such as navigation of autonomous agents. Even though CNN-based SLAM algorithms are a growing area of research there is a notable absence of a comprehensive presentation and examination of adversarial attacks targeting CNN-based feature detectors, as part of a SLAM system. Our work introduces black-box adversarial perturbations applied to the RGB images fed into the GCN-SLAM algorithm. Our findings on the TUM dataset [30] reveal that even attacks of moderate scale can lead to tracking failure in as many as 76% of the frames. Moreover, our experiments highlight the catastrophic impact of attacking depth instead of RGB input images on the SLAM system.

Bingxiang He, Wenbin Zhang, Jiaxi Song et al.

Preference learning is critical for aligning large language models (LLMs) with human values, yet its success hinges on high-quality datasets comprising three core components: Preference \textbf{A}nnotations, \textbf{I}nstructions, and \textbf{R}esponse Pairs. Current approaches conflate these components, obscuring their individual impacts and hindering systematic optimization. In this work, we propose \textbf{AIR}, a component-wise analysis framework that systematically isolates and optimizes each component while evaluating their synergistic effects. Through rigorous experimentation, AIR reveals actionable principles: annotation simplicity (point-wise generative scoring), instruction inference stability (variance-based filtering across LLMs), and response pair quality (moderate margins + high absolute scores). When combined, these principles yield +5.3 average gains over baseline method, even with only 14k high-quality pairs. Our work shifts preference dataset design from ad hoc scaling to component-aware optimization, offering a blueprint for efficient, reproducible alignment.

Bowen Sun, Shibao Zheng

The blooming diffusion probabilistic models (DPMs) have garnered significant interest due to their impressive performance and the elegant inspiration they draw from physics. While earlier DPMs relied upon the Markovian assumption, recent methods based on differential equations have been rapidly applied to enhance the efficiency and capabilities of these models. However, a theoretical interpretation encapsulating these diverse algorithms is insufficient yet pressingly required to guide further development of DPMs. In response to this need, we present FreeFlow, a framework that provides a thorough explanation of the diffusion formula as time-dependent optimal transport, where the evolutionary pattern of probability density is given by the gradient flows of a functional defined in Wasserstein space. Crucially, our framework necessitates a unified description that not only clarifies the subtle mechanism of DPMs but also indicates the roots of some defects through creative involvement of Lagrangian and Eulerian views to understand the evolution of probability flow. We particularly demonstrate that the core equation of FreeFlow condenses all stochastic and deterministic DPMs into a single case, showcasing the expansibility of our method. Furthermore, the Riemannian geometry employed in our work has the potential to bridge broader subjects in mathematics, which enable the involvement of more profound tools for the establishment of more outstanding and generalized models in the future.

Congcong Wu, Xiangyun Gao, Xueyong Liu et al.

The set-union knapsack problem (SUKP) is a constrained composed optimization problem. It is more difficulty for solving because values and weights depend on items and elements respectively. In this paper, we present two self-adjusting optimization algorithms for approximating SUKP from items and elements perspective respectively. By analyzing the dynamic characters in the SUKP, we design two types of self-adjusting repair and optimization operators that are based on the different loading process. We use the novel teaching-learning-based optimization algorithm (TLBO) to design a general discrete framework (DTLBO) suitable for these two types of operators. In addition, we introduce elite opposite search and natural selection mechanism into DTLBO to furtherly improve the performance of the algorithm from the perspective of population. Finally, we performed experimental comparisons on benchmark sets to verify the effectiveness of the proposed algorithm. The experimental results show that the item-based self-adjusting optimization algorithm I-DTLBO is outstanding, and the algorithm is superior to the other swarm intelligence methods for solving SUKP. IDTLBO algorithm reaches the upper boundary of the current swarm intelligence algorithms for solving SUKP in 10 instances, and gotten new upper boundary in 15 instances. The algorithm E-DTLBO based on element loading only perform slightly better on small and middle data sets, but worse on large-scale instances. It shows that element-based design is not suitable for solving SUKP.

Xuhong Wang, Ding Lyu, Mengjian Li et al.

Limited by the time complexity of querying k-hop neighbors in a graph database, most graph algorithms cannot be deployed online and execute millisecond-level inference. This problem dramatically limits the potential of applying graph algorithms in certain areas, such as financial fraud detection. Therefore, we propose Asynchronous Propagation Attention Network, an asynchronous continuous time dynamic graph algorithm for real-time temporal graph embedding. Traditional graph models usually execute two serial operations: first graph computation and then model inference. We decouple model inference and graph computation step so that the heavy graph query operations will not damage the speed of model inference. Extensive experiments demonstrate that the proposed method can achieve competitive performance and 8.7 times inference speed improvement in the meantime.