James Caverlee

Zhitong Chen, Kai Yin, Weifeng Zhang et al.

Disasters cause severe societal impacts, demanding rapid coordination of heterogeneous AI tools, from satellite analysis to flood prediction and damage assessment, into coherent multi-step workflows. As LLMs increasingly serve as orchestrators of such pipelines, effective coordination requires more than selecting semantically plausible tools: LLMs must generate executable workflows with correct parameter binding and dependency propagation. We introduce DisasterBench, a benchmark for evaluating structured multi-agent planning over semantically similar but operationally distinct disaster-response tools. To enable step-level failure attribution, we further propose First-Point-of-Failure (FPoF), which localizes the earliest root cause in a predicted workflow, separating primary errors from downstream cascading effects. Our evaluation reveals three findings: planning method effectiveness depends strongly on model capacity; tool mismatch and parameter-binding errors dominate first failures, revealing semantic grounding and execution consistency as distinct bottlenecks; and verbose intermediate reasoning can create instruction clash with structured output requirements, disrupting plan generation. Together, these findings highlight a fundamental gap between semantic reasoning and execution-grounded coordination, underscoring the need for planning frameworks that jointly model semantic intent, execution constraints, and workflow consistency. Code, data, and evaluation resources are available at: https://github.com/TamuChen18/DisasterBench_Open

Yun He, Xue Feng, Cheng Cheng et al.

In many personalized recommendation scenarios, the generalization ability of a target task can be improved via learning with additional auxiliary tasks alongside this target task on a multi-task network. However, this method often suffers from a serious optimization imbalance problem. On the one hand, one or more auxiliary tasks might have a larger influence than the target task and even dominate the network weights, resulting in worse recommendation accuracy for the target task. On the other hand, the influence of one or more auxiliary tasks might be too weak to assist the target task. More challenging is that this imbalance dynamically changes throughout the training process and varies across the parts of the same network. We propose a new method: MetaBalance to balance auxiliary losses via directly manipulating their gradients w.r.t the shared parameters in the multi-task network. Specifically, in each training iteration and adaptively for each part of the network, the gradient of an auxiliary loss is carefully reduced or enlarged to have a closer magnitude to the gradient of the target loss, preventing auxiliary tasks from being so strong that dominate the target task or too weak to help the target task. Moreover, the proximity between the gradient magnitudes can be flexibly adjusted to adapt MetaBalance to different scenarios. The experiments show that our proposed method achieves a significant improvement of 8.34% in terms of NDCG@10 upon the strongest baseline on two real-world datasets. The code of our approach can be found at here: https://github.com/facebookresearch/MetaBalance

Xiangjue Dong, Ziwei Zhu, Zhuoer Wang et al. · amazon-science, microsoft-research

Pre-trained Language Models are widely used in many important real-world applications. However, recent studies show that these models can encode social biases from large pre-training corpora and even amplify biases in downstream applications. To address this challenge, we propose Co$^2$PT, an efficient and effective debias-while-prompt tuning method for mitigating biases via counterfactual contrastive prompt tuning on downstream tasks. Our experiments conducted on three extrinsic bias benchmarks demonstrate the effectiveness of Co$^2$PT on bias mitigation during the prompt tuning process and its adaptability to existing upstream debiased language models. These findings indicate the strength of Co$^2$PT and provide promising avenues for further enhancement in bias mitigation on downstream tasks.

Shubham Parashar, Shurui Gui, Xiner Li et al.

We aim to improve the reasoning capabilities of language models via reinforcement learning (RL). Recent RL post-trained models like DeepSeek-R1 have demonstrated reasoning abilities on mathematical and coding tasks. However, prior studies suggest that using RL alone to improve reasoning on inherently difficult tasks is less effective. Here, we draw inspiration from curriculum learning and propose to schedule tasks from easy to hard (E2H), allowing LLMs to build reasoning skills gradually. Our method is termed E2H Reasoner. Empirically, we observe that, although easy tasks are important initially, fading them out through appropriate scheduling is essential in preventing overfitting. Theoretically, we establish convergence guarantees for E2H Reasoner within an approximate policy iteration framework. We derive finite-sample complexity bounds and show that when tasks are appropriately decomposed and conditioned, learning through curriculum stages requires fewer total samples than direct learning. Experiments across multiple domains show that E2H Reasoner significantly improves the reasoning ability of small LLMs (1.5B to 3B), which otherwise struggle when trained with vanilla RL alone, highlighting the effectiveness of our method. Our code can be found on https://github.com/divelab/E2H-Reasoning.

Zhuoer Wang, Leonardo F. R. Ribeiro, Alexandros Papangelis et al. · amazon-science, microsoft-research

API call generation is the cornerstone of large language models' tool-using ability that provides access to the larger world. However, existing supervised and in-context learning approaches suffer from high training costs, poor data efficiency, and generated API calls that can be unfaithful to the API documentation and the user's request. To address these limitations, we propose an output-side optimization approach called FANTASE. Two of the unique contributions of FANTASE are its State-Tracked Constrained Decoding (SCD) and Reranking components. SCD dynamically incorporates appropriate API constraints in the form of Token Search Trie for efficient and guaranteed generation faithfulness with respect to the API documentation. The Reranking component efficiently brings in the supervised signal by leveraging a lightweight model as the discriminator to rerank the beam-searched candidate generations of the large language model. We demonstrate the superior performance of FANTASE in API call generation accuracy, inference efficiency, and context efficiency with DSTC8 and API Bank datasets.

Zhuoer Wang, Yicheng Wang, Ziwei Zhu et al. · amazon-science, microsoft-research

Question generation is a widely used data augmentation approach with extensive applications, and extracting qualified candidate answers from context passages is a critical step for most question generation systems. However, existing methods for candidate answer extraction are reliant on linguistic rules or annotated data that face the partial annotation issue and challenges in generalization. To overcome these limitations, we propose a novel unsupervised candidate answer extraction approach that leverages the inherent structure of context passages through a Differentiable Masker-Reconstructor (DMR) Model with the enforcement of self-consistency for picking up salient information tokens. We curated two datasets with exhaustively-annotated answers and benchmark a comprehensive set of supervised and unsupervised candidate answer extraction methods. We demonstrate the effectiveness of the DMR model by showing its performance is superior among unsupervised methods and comparable to supervised methods.

Xiangjue Dong, Yibo Wang, Philip S. Yu et al.

Large Language Models (LLMs) can generate biased and toxic responses. Yet most prior work on LLM gender bias evaluation requires predefined gender-related phrases or gender stereotypes, which are challenging to be comprehensively collected and are limited to explicit bias evaluation. In addition, we believe that instances devoid of gender-related language or explicit stereotypes in inputs can still induce gender bias in LLMs. Thus, in this work, we propose a conditional text generation mechanism without the need for predefined gender phrases and stereotypes. This approach employs three types of inputs generated through three distinct strategies to probe LLMs, aiming to show evidence of explicit and implicit gender biases in LLMs. We also utilize explicit and implicit evaluation metrics to evaluate gender bias in LLMs under different strategies. Our experiments demonstrate that an increased model size does not consistently lead to enhanced fairness and all tested LLMs exhibit explicit and/or implicit gender bias, even when explicit gender stereotypes are absent in the inputs.

Shubham Parashar, Atharv Chagi, Jacob Helwig et al.

We aim to improve the reasoning capabilities of diffusion language models (DLMs). While SFT is a popular post-training recipe for autoregressive models, its use in DLMs faces challenges and can even hurt performance, though the underlying causes remain understudied. Our analysis reveals that vanilla SFT overlooks learnability, namely what and when tokens are learned. Specifically, rare tokens are difficult to learn when most of the input is masked, whereas it is straightforward and thus of little value to learn common tokens when most of the input is unmasked. Motivated by our analysis, we propose LIFT, an efficient SFT-based post-training algorithm for DLMs. LIFT learns easy tokens when most of the input is masked and hard tokens when more context is available, thus aligning the training with the information available at different diffusion time steps. Our results show that LIFT outperforms existing SFT baselines across six reasoning benchmarks, achieving up to a 3x relative gain on AIME'24 and AIME'25. Our code is publicly available at https://github.com/divelab/LIFT.

Xiangjue Dong, Jiaying Lu, Jianling Wang et al.

Question Generation (QG) is a fundamental NLP task for many downstream applications. Recent studies on open-book QG, where supportive answer-context pairs are provided to models, have achieved promising progress. However, generating natural questions under a more practical closed-book setting that lacks these supporting documents still remains a challenge. In this work, we propose a new QG model for this closed-book setting that is designed to better understand the semantics of long-form abstractive answers and store more information in its parameters through contrastive learning and an answer reconstruction module. Through experiments, we validate the proposed QG model on both public datasets and a new WikiCQA dataset. Empirical results show that the proposed QG model outperforms baselines in both automatic evaluation and human evaluation. In addition, we show how to leverage the proposed model to improve existing question-answering systems. These results further indicate the effectiveness of our QG model for enhancing closed-book question-answering tasks.

Xiangjue Dong, Yun He, Ziwei Zhu et al.

A key component of modern conversational systems is the Dialogue State Tracker (or DST), which models a user's goals and needs. Toward building more robust and reliable DSTs, we introduce a prompt-based learning approach to automatically generate effective adversarial examples to probe DST models. Two key characteristics of this approach are: (i) it only needs the output of the DST with no need for model parameters, and (ii) it can learn to generate natural language utterances that can target any DST. Through experiments over state-of-the-art DSTs, the proposed framework leads to the greatest reduction in accuracy and the best attack success rate while maintaining good fluency and a low perturbation ratio. We also show how much the generated adversarial examples can bolster a DST through adversarial training. These results indicate the strength of prompt-based attacks on DSTs and leave open avenues for continued refinement.

Haoran Liu, Bokun Wang, Jianling Wang et al.

As powerful tools for representation learning on graphs, graph neural networks (GNNs) have played an important role in applications including social networks, recommendation systems, and online web services. However, GNNs have been shown to be vulnerable to adversarial attacks, which can significantly degrade their effectiveness. Recent state-of-the-art approaches in adversarial attacks rely on gradient-based meta-learning to selectively perturb a single edge with the highest attack score until they reach the budget constraint. While effective in identifying vulnerable links, these methods are plagued by high computational costs. By leveraging continuous relaxation and parameterization of the graph structure, we propose a novel attack method called Differentiable Graph Attack (DGA) to efficiently generate effective attacks and meanwhile eliminate the need for costly retraining. Compared to the state-of-the-art, DGA achieves nearly equivalent attack performance with 6 times less training time and 11 times smaller GPU memory footprint on different benchmark datasets. Additionally, we provide extensive experimental analyses of the transferability of the DGA among different graph models, as well as its robustness against widely-used defense mechanisms.

Yibo Wang, Xiangjue Dong, James Caverlee et al.

Language models can be manipulated by adversarial attacks, which introduce subtle perturbations to input data. While recent attack methods can achieve a relatively high attack success rate (ASR), we've observed that the generated adversarial examples have a different data distribution compared with the original examples. Specifically, these adversarial examples exhibit reduced confidence levels and greater divergence from the training data distribution. Consequently, they are easy to detect using straightforward detection methods, diminishing the efficacy of such attacks. To address this issue, we propose a Distribution-Aware Adversarial Attack ($DA^3$) method. $DA^3$ considers the distribution shifts of adversarial examples to improve attacks' effectiveness under detection methods. We further design a novel evaluation metric, the Non-detectable Attack Success Rate (NASR), which integrates both ASR and detectability for the attack task. We conduct experiments on four widely used datasets to validate the attack effectiveness and transferability of adversarial examples generated by $DA^3$ against both the white-box BERT-base and RoBERTa-base models and the black-box LLaMA2-7b model.

Zhitong Chen, Kai Yin, Xiangjue Dong et al.

Accurate question answering (QA) in disaster management requires reasoning over uncertain and conflicting information, a setting poorly captured by existing benchmarks built on clean evidence. We introduce DisastQA, a large-scale benchmark of 3,000 rigorously verified questions (2,000 multiple-choice and 1,000 open-ended) spanning eight disaster types. The benchmark is constructed via a human-LLM collaboration pipeline with stratified sampling to ensure balanced coverage. Models are evaluated under varying evidence conditions, from closed-book to noisy evidence integration, enabling separation of internal knowledge from reasoning under imperfect information. For open-ended QA, we propose a human-verified keypoint-based evaluation protocol emphasizing factual completeness over verbosity. Experiments with 20 models reveal substantial divergences from general-purpose leaderboards such as MMLU-Pro. While recent open-weight models approach proprietary systems in clean settings, performance degrades sharply under realistic noise, exposing critical reliability gaps for disaster response. All code, data, and evaluation resources are available at https://github.com/TamuChen18/DisastQA_open.

Jason Kim, Maria Teleki, James Caverlee

Prompting is central to interaction with AI systems, yet many users struggle to explore alternative directions, articulate creative intent, or understand how variations in prompts shape model outputs. We introduce prompt recommender systems (PRS) as an interaction approach that supports exploration, suggesting contextually relevant follow-up prompts. We present PromptHelper, a PRS prototype integrated into an AI chatbot that surfaces semantically diverse prompt suggestions while users work on real writing tasks. We evaluate PromptHelper in a 2x2 fully within-subjects study (N=32) across creative and academic writing tasks. Results show that PromptHelper significantly increases users' perceived exploration and expressiveness without increasing cognitive workload. Qualitative findings illustrate how prompt recommendations help users branch into new directions, overcome uncertainty about what to ask next, and better articulate their intent. We discuss implications for designing AI interfaces that scaffold exploratory interaction while preserving user agency, and release open-source resources to support research on prompt recommendation.

Tian Liu, Anwesha Basu, James Caverlee et al.

Semi-supervised few-shot learning (SSFSL) formulates real-world applications like ''auto-annotation'', as it aims to learn a model over a few labeled and abundant unlabeled examples to annotate the unlabeled ones. Despite the availability of powerful open-source Vision-Language Models (VLMs) and their pretraining data, the SSFSL literature largely neglects these open-source resources. In contrast, the related area few-shot learning (FSL) has already exploited them to boost performance. Arguably, to achieve auto-annotation in the real world, SSFSL should leverage such open-source resources. To this end, we start by applying established SSL methods to finetune a VLM. Counterintuitively, they significantly underperform FSL baselines. Our in-depth analysis reveals the root cause: VLMs produce rather ''flat'' distributions of softmax probabilities. This results in zero utilization of unlabeled data and weak supervision signals. We address this issue with embarrassingly simple techniques: classifier initialization and temperature tuning. They jointly increase the confidence scores of pseudo-labels, improving the utilization rate of unlabeled data, and strengthening supervision signals. Building on this, we propose: Stage-Wise Finetuning with Temperature Tuning (SWIFT), which enables existing SSL methods to effectively finetune a VLM on limited labeled data, abundant unlabeled data, and task-relevant but noisy data retrieved from the VLM's pretraining set. Extensive experiments on five SSFSL benchmarks show that SWIFT outperforms recent FSL and SSL methods by $\sim$5 accuracy points. SWIFT even rivals supervised learning, which finetunes VLMs with the unlabeled data being labeled with ground truth!

Weizhi Zhang, Yuanchen Bei, Liangwei Yang et al. · tsinghua

Cold-start problem is one of the long-standing challenges in recommender systems, focusing on accurately modeling new or interaction-limited users or items to provide better recommendations. Due to the diversification of internet platforms and the exponential growth of users and items, the importance of cold-start recommendation (CSR) is becoming increasingly evident. At the same time, large language models (LLMs) have achieved tremendous success and possess strong capabilities in modeling user and item information, providing new potential for cold-start recommendations. However, the research community on CSR still lacks a comprehensive review and reflection in this field. Based on this, in this paper, we stand in the context of the era of large language models and provide a comprehensive review and discussion on the roadmap, related literature, and future directions of CSR. Specifically, we have conducted an exploration of the development path of how existing CSR utilizes information, from content features, graph relations, and domain information, to the world knowledge possessed by large language models, aiming to provide new insights for both the research and industrial communities on CSR. Related resources of cold-start recommendations are collected and continuously updated for the community in https://github.com/YuanchenBei/Awesome-Cold-Start-Recommendation.

Millennium Bismay, Xiangjue Dong, James Caverlee

This paper presents ReasoningRec, a reasoning-based recommendation framework that leverages Large Language Models (LLMs) to bridge the gap between recommendations and human-interpretable explanations. In contrast to conventional recommendation systems that rely on implicit user-item interactions, ReasoningRec employs LLMs to model users and items, focusing on preferences, aversions, and explanatory reasoning. The framework utilizes a larger LLM to generate synthetic explanations for user preferences, subsequently used to fine-tune a smaller LLM for enhanced recommendation accuracy and human-interpretable explanation. Our experimental study investigates the impact of reasoning and contextual information on personalized recommendations, revealing that the quality of contextual and personalized data significantly influences the LLM's capacity to generate plausible explanations. Empirical evaluations demonstrate that ReasoningRec surpasses state-of-the-art methods by up to 12.5\% in recommendation prediction while concurrently providing human-intelligible explanations. The code is available here: https://github.com/millenniumbismay/reasoningrec.

Kai Yin, Xiangjue Dong, Chengkai Liu et al.

Effective disaster management requires timely access to accurate and contextually relevant information. Existing Information Retrieval (IR) benchmarks, however, focus primarily on general or specialized domains, such as medicine or finance, neglecting the unique linguistic complexity and diverse information needs encountered in disaster management scenarios. To bridge this gap, we introduce DisastIR, the first comprehensive IR evaluation benchmark specifically tailored for disaster management. DisastIR comprises 9,600 diverse user queries and more than 1.3 million labeled query-passage pairs, covering 48 distinct retrieval tasks derived from six search intents and eight general disaster categories that include 301 specific event types. Our evaluations of 30 state-of-the-art retrieval models demonstrate significant performance variances across tasks, with no single model excelling universally. Furthermore, comparative analyses reveal significant performance gaps between general-domain and disaster management-specific tasks, highlighting the necessity of disaster management-specific benchmarks for guiding IR model selection to support effective decision-making in disaster management scenarios. All source codes and DisastIR are available at https://github.com/KaiYin97/Disaster_IR.

Jinhao Pan, Ziwei Zhu, Jianling Wang et al.

Collaborative filtering (CF) based recommendations suffer from mainstream bias -- where mainstream users are favored over niche users, leading to poor recommendation quality for many long-tail users. In this paper, we identify two root causes of this mainstream bias: (i) discrepancy modeling, whereby CF algorithms focus on modeling mainstream users while neglecting niche users with unique preferences; and (ii) unsynchronized learning, where niche users require more training epochs than mainstream users to reach peak performance. Targeting these causes, we propose a novel end-To-end Adaptive Local Learning (TALL) framework to provide high-quality recommendations to both mainstream and niche users. TALL uses a loss-driven Mixture-of-Experts module to adaptively ensemble experts to provide customized local models for different users. Further, it contains an adaptive weight module to synchronize the learning paces of different users by dynamically adjusting weights in the loss. Extensive experiments demonstrate the state-of-the-art performance of the proposed model. Code and data are provided at \url{https://github.com/JP-25/end-To-end-Adaptive-Local-Leanring-TALL-}

Shree Harsha Bokkahalli Satish, Christoph Minixhofer, Maria Teleki et al.

Speech Large Language Models (SpeechLLMs) process spoken input directly, retaining cues such as accent and perceived gender that were previously removed in cascaded pipelines. This introduces speaker identity dependent variation in responses. We present a large-scale intersectional evaluation of accent and gender bias in three SpeechLLMs using 2,880 controlled interactions across six English accents and two gender presentations, keeping linguistic content constant through voice cloning. Using pointwise LLM-judge ratings, pairwise comparisons, and Best-Worst Scaling with human validation, we detect consistent disparities. Eastern European-accented speech receives lower helpfulness scores, particularly for female-presenting voices. The bias is implicit: responses remain polite but differ in helpfulness. While LLM judges capture the directional trend of these biases, human evaluators exhibit significantly higher sensitivity, uncovering sharper intersectional disparities.

Mahsa Valizadeh, Rui Tuo, James Caverlee

Generative Adversarial Networks (GANs) are proficient at generating synthetic data but continue to suffer from mode collapse, where the generator produces a narrow range of outputs that fool the discriminator but fail to capture the full data distribution. This limitation is particularly problematic, as generative models are increasingly deployed in real-world applications that demand both diversity and uncertainty awareness. In response, we introduce BI-DCGAN, a Bayesian extension of DCGAN that incorporates model uncertainty into the generative process while maintaining computational efficiency. BI-DCGAN integrates Bayes by Backprop to learn a distribution over network weights and employs mean-field variational inference to efficiently approximate the posterior distribution during GAN training. We establishes the first theoretical proof, based on covariance matrix analysis, that Bayesian modeling enhances sample diversity in GANs. We validate this theoretical result through extensive experiments on standard generative benchmarks, demonstrating that BI-DCGAN produces more diverse and robust outputs than conventional DCGANs, while maintaining training efficiency. These findings position BI-DCGAN as a scalable and timely solution for applications where both diversity and uncertainty are critical, and where modern alternatives like diffusion models remain too resource-intensive.

Kai Yin, Xiangjue Dong, Chengkai Liu et al.

Effective and efficient access to relevant information is essential for disaster management. However, no retrieval model is specialized for disaster management, and existing general-domain models fail to handle the varied search intents inherent to disaster management scenarios, resulting in inconsistent and unreliable performance. To this end, we introduce DMRetriever, the first series of dense retrieval models (33M to 7.6B) tailored for this domain. It is trained through a novel three-stage framework of bidirectional attention adaptation, unsupervised contrastive pre-training, and difficulty-aware progressive instruction fine-tuning, using high-quality data generated through an advanced data refinement pipeline. Comprehensive experiments demonstrate that DMRetriever achieves state-of-the-art (SOTA) performance across all six search intents at every model scale. Moreover, DMRetriever is highly parameter-efficient, with 596M model outperforming baselines over 13.3 X larger and 33M model exceeding baselines with only 7.6% of their parameters. All codes, data, and checkpoints are available at https://github.com/KaiYin97/DMRETRIEVER

Maria Teleki, Sai Janjur, Haoran Liu et al.

Disfluencies -- such as "um," "uh," interjections, parentheticals, and edited statements -- remain a persistent challenge for speech-driven systems, degrading accuracy in command interpretation, summarization, and conversational agents. We introduce DRES (Disfluency Removal Evaluation Suite), a controlled text-level benchmark that establishes a reproducible semantic upper bound for this task. DRES builds on human-annotated Switchboard transcripts, isolating disfluency removal from ASR errors and acoustic variability. We systematically evaluate proprietary and open-source LLMs across scales, prompting strategies, and architectures. Our results reveal that (i) simple segmentation consistently improves performance, even for long-context models; (ii) reasoning-oriented models tend to over-delete fluent tokens; and (iii) fine-tuning achieves near state-of-the-art precision and recall but harms generalization abilities. We further present a set of LLM-specific error modes and offer nine practical recommendations (R1-R9) for deploying disfluency removal in speech-driven pipelines. DRES provides a reproducible, model-agnostic foundation for advancing robust spoken-language systems.

Jian Wu, Rajal Nivargi, Sree Sai Teja Lanka et al.

In recent years, significant effort has been invested verifying the reproducibility and robustness of research claims in social and behavioral sciences (SBS), much of which has involved resource-intensive replication projects. In this paper, we investigate prediction of the reproducibility of SBS papers using machine learning methods based on a set of features. We propose a framework that extracts five types of features from scholarly work that can be used to support assessments of reproducibility of published research claims. Bibliometric features, venue features, and author features are collected from public APIs or extracted using open source machine learning libraries with customized parsers. Statistical features, such as p-values, are extracted by recognizing patterns in the body text. Semantic features, such as funding information, are obtained from public APIs or are extracted using natural language processing models. We analyze pairwise correlations between individual features and their importance for predicting a set of human-assessed ground truth labels. In doing so, we identify a subset of 9 top features that play relatively more important roles in predicting the reproducibility of SBS papers in our corpus. Results are verified by comparing performances of 10 supervised predictive classifiers trained on different sets of features.

Yun He, Yin Zhang, Weiwen Liu et al.

User-generated item lists are popular on many platforms. Examples include video-based playlists on YouTube, image-based lists (or"boards") on Pinterest, book-based lists on Goodreads, and answer-based lists on question-answer forums like Zhihu. As users create these lists, a common challenge is in identifying what items to curate next. Some lists are organized around particular genres or topics, while others are seemingly incoherent, reflecting individual preferences for what items belong together. Furthermore, this heterogeneity in item consistency may vary from platform to platform, and from sub-community to sub-community. Hence, this paper proposes a generalizable approach for user-generated item list continuation. Complementary to methods that exploit specific content patterns (e.g., as in song-based playlists that rely on audio features), the proposed approach models the consistency of item lists based on human curation patterns, and so can be deployed across a wide range of varying item types (e.g., videos, images, books). A key contribution is in intelligently combining two preference models via a novel consistency-aware gating network - a general user preference model that captures a user's overall interests, and a current preference priority model that captures a user's current (as of the most recent item) interests. In this way, the proposed consistency-aware recommender can dynamically adapt as user preferences evolve. Evaluation over four datasets(of songs, books, and answers) confirms these observations and demonstrates the effectiveness of the proposed model versus state-of-the-art alternatives. Further, all code and data are available at https://github.com/heyunh2015/ListContinuation_WSDM2020.

Yun He, Jianling Wang, Wei Niu et al.

User-generated item lists are a popular feature of many different platforms. Examples include lists of books on Goodreads, playlists on Spotify and YouTube, collections of images on Pinterest, and lists of answers on question-answer sites like Zhihu. Recommending item lists is critical for increasing user engagement and connecting users to new items, but many approaches are designed for the item-based recommendation, without careful consideration of the complex relationships between items and lists. Hence, in this paper, we propose a novel user-generated list recommendation model called AttList. Two unique features of AttList are careful modeling of (i) hierarchical user preference, which aggregates items to characterize the list that they belong to, and then aggregates these lists to estimate the user preference, naturally fitting into the hierarchical structure of item lists; and (ii) item and list consistency, through a novel self-attentive aggregation layer designed for capturing the consistency of neighboring items and lists to better model user preference. Through experiments over three real-world datasets reflecting different kinds of user-generated item lists, we find that AttList results in significant improvements in NDCG, Precision@k, and Recall@k versus a suite of state-of-the-art baselines. Furthermore, all code and data are available at https://github.com/heyunh2015/AttList.

Yun He, Haochen Chen, Ziwei Zhu et al.

We propose PsiRec, a novel user preference propagation recommender that incorporates pseudo-implicit feedback for enriching the original sparse implicit feedback dataset. Three of the unique characteristics of PsiRec are: (i) it views user-item interactions as a bipartite graph and models pseudo-implicit feedback from this perspective; (ii) its random walks-based approach extracts graph structure information from this bipartite graph, toward estimating pseudo-implicit feedback; and (iii) it adopts a Skip-gram inspired measure of confidence in pseudo-implicit feedback that captures the pointwise mutual information between users and items. This pseudo-implicit feedback is ultimately incorporated into a new latent factor model to estimate user preference in cases of extreme sparsity. PsiRec results in improvements of 21.5% and 22.7% in terms of Precision@10 and Recall@10 over state-of-the-art Collaborative Denoising Auto-Encoders. Our implementation is available at https://github.com/heyunh2015/PsiRecICDM2018.

Noveen Sachdeva, Benjamin Coleman, Wang-Cheng Kang et al.

The training of large language models (LLMs) is expensive. In this paper, we study data-efficient approaches for pre-training LLMs, i.e., techniques that aim to optimize the Pareto frontier of model quality and training resource/data consumption. We seek to understand the tradeoffs associated with data selection routines based on (i) expensive-to-compute data-quality estimates, and (ii) maximization of coverage and diversity-based measures in the feature space. Our first technique, Ask-LLM, leverages the zero-shot reasoning capabilities of instruction-tuned LLMs to directly assess the quality of a training example. To target coverage, we propose Density sampling, which models the data distribution to select a diverse sample. In our comparison of 19 samplers, involving hundreds of evaluation tasks and pre-training runs, we find that Ask-LLM and Density are the best methods in their respective categories. Coverage sampling can recover the performance of the full data, while models trained on Ask-LLM data consistently outperform full-data training -- even when we reject 90% of the original dataset, while converging up to 70% faster.

Anna Seo Gyeong Choi, Maria Teleki, James Caverlee et al.

Automatic speech recognition (ASR) evaluation compares system output to ground truth transcripts, with Word Error Rate (WER) quantifying the distance between them. But ground truth transcripts are not discovered - they are produced by human annotators following conventions that encode normative assumptions about which speech features matter. Different conventions (verbatim, non-verbatim, legal) produce different transcripts of identical speech and judge the same ASR output differently. This paper argues that reference monism - enforcing a single transcription convention as ground truth - commits epistemic injustice. Speakers with aphasia, whose speech includes clinically meaningful disfluencies, are systematically disadvantaged when evaluated against "clean" references that treat those disfluencies as errors. The harm is not merely differential performance, but that evaluative infrastructure lacks interpretive resources to recognize their contributions as legitimate. We develop a philosophical framework introducing the hermeneutical gap, formalize Epistemic Injustice Distance (EID) to measure reference monism's cost, and demonstrate empirically using AphasiaBank that WER varies depending on which convention defines ground truth. We propose WER-Range: reporting performance across legitimate conventions rather than assuming a single correct answer.

Shubham Parashar, Zhiqiu Lin, Tian Liu et al.

Vision-language models (VLMs) excel in zero-shot recognition but their performance varies greatly across different visual concepts. For example, although CLIP achieves impressive accuracy on ImageNet (60-80%), its performance drops below 10% for more than ten concepts like night snake, presumably due to their limited presence in the pretraining data. However, measuring the frequency of concepts in VLMs' large-scale datasets is challenging. We address this by using large language models (LLMs) to count the number of pretraining texts that contain synonyms of these concepts. Our analysis confirms that popular datasets, such as LAION, exhibit a long-tailed concept distribution, yielding biased performance in VLMs. We also find that downstream applications of VLMs, including visual chatbots (e.g., GPT-4V) and text-to-image models (e.g., Stable Diffusion), often fail to recognize or generate images of rare concepts identified by our method. To mitigate the imbalanced performance of zero-shot VLMs, we propose REtrieval-Augmented Learning (REAL). First, instead of prompting VLMs using the original class names, REAL uses their most frequent synonyms found in pretraining texts. This simple change already outperforms costly human-engineered and LLM-enriched prompts over nine benchmark datasets. Second, REAL trains a linear classifier on a small yet balanced set of pretraining data retrieved using concept synonyms. REAL surpasses the previous zero-shot SOTA, using 400x less storage and 10,000x less training time!

Rui Yang, Haoran Liu, Edison Marrese-Taylor et al.

Large language models (LLMs) have demonstrated impressive generative capabilities with the potential to innovate in medicine. However, the application of LLMs in real clinical settings remains challenging due to the lack of factual consistency in the generated content. In this work, we develop an augmented LLM framework, KG-Rank, which leverages a medical knowledge graph (KG) along with ranking and re-ranking techniques, to improve the factuality of long-form question answering (QA) in the medical domain. Specifically, when receiving a question, KG-Rank automatically identifies medical entities within the question and retrieves the related triples from the medical KG to gather factual information. Subsequently, KG-Rank innovatively applies multiple ranking techniques to refine the ordering of these triples, providing more relevant and precise information for LLM inference. To the best of our knowledge, KG-Rank is the first application of KG combined with ranking models in medical QA specifically for generating long answers. Evaluation on four selected medical QA datasets demonstrates that KG-Rank achieves an improvement of over 18% in ROUGE-L score. Additionally, we extend KG-Rank to open domains, including law, business, music, and history, where it realizes a 14% improvement in ROUGE-L score, indicating the effectiveness and great potential of KG-Rank.

Xiangjue Dong, Yibo Wang, Philip S. Yu et al.

Large Language Models (LLMs) can generate biased responses. Yet previous direct probing techniques contain either gender mentions or predefined gender stereotypes, which are challenging to comprehensively collect. Hence, we propose an indirect probing framework based on conditional generation. This approach aims to induce LLMs to disclose their gender bias even without explicit gender or stereotype mentions. We explore three distinct strategies to disclose explicit and implicit gender bias in LLMs. Our experiments demonstrate that all tested LLMs exhibit explicit and/or implicit gender bias, even when gender stereotypes are not present in the inputs. In addition, an increased model size or model alignment amplifies bias in most cases. Furthermore, we investigate three methods to mitigate bias in LLMs via Hyperparameter Tuning, Instruction Guiding, and Debias Tuning. Remarkably, these methods prove effective even in the absence of explicit genders or stereotypes.

Shubham Parashar, Blake Olson, Sambhav Khurana et al.

We examine the reasoning and planning capabilities of large language models (LLMs) in solving complex tasks. Recent advances in inference-time techniques demonstrate the potential to enhance LLM reasoning without additional training by exploring intermediate steps during inference. Notably, OpenAI's o1 model shows promising performance through its novel use of multi-step reasoning and verification. Here, we explore how scaling inference-time techniques can improve reasoning and planning, focusing on understanding the tradeoff between computational cost and performance. To this end, we construct a comprehensive benchmark, known as Sys2Bench, and perform extensive experiments evaluating existing inference-time techniques on eleven diverse tasks across five categories, including arithmetic reasoning, logical reasoning, common sense reasoning, algorithmic reasoning, and planning. Our findings indicate that simply scaling inference-time computation has limitations, as no single inference-time technique consistently performs well across all reasoning and planning tasks.

Xiangjue Dong, Maria Teleki, James Caverlee

Techniques that enhance inference through increased computation at test-time have recently gained attention. In this survey, we investigate the current state of LLM Inference-Time Self-Improvement from three different perspectives: Independent Self-improvement, focusing on enhancements via decoding or sampling methods; Context-Aware Self-Improvement, leveraging additional context or datastore; and Model-Aided Self-Improvement, achieving improvement through model collaboration. We provide a comprehensive review of recent relevant studies, contribute an in-depth taxonomy, and discuss challenges and limitations, offering insights for future research.

Hongyi Ling, Shubham Parashar, Sambhav Khurana et al.

We consider enhancing large language models (LLMs) for complex planning tasks. While existing methods allow LLMs to explore intermediate steps to make plans, they either depend on unreliable self-verification or external verifiers to evaluate these steps, which demand significant data and computations. Here, we propose automated heuristics discovery (AutoHD), a novel approach that enables LLMs to explicitly generate heuristic functions to guide inference-time search, allowing accurate evaluation of intermediate states. These heuristic functions are further refined through a heuristic evolution process, improving their robustness and effectiveness. Our proposed method requires no additional model training or fine-tuning, and the explicit definition of heuristic functions generated by the LLMs provides interpretability and insights into the reasoning process. Extensive experiments across diverse benchmarks demonstrate significant gains over multiple baselines, including nearly twice the accuracy on some datasets, establishing our approach as a reliable and interpretable solution for complex planning tasks.

Yingjian Chen, Haoran Liu, Yinhong Liu et al.

Large language models (LLMs) are widely used, but they often generate subtle factual errors, especially in long-form text. These errors are fatal in some specialized domains such as medicine. Existing fact-checking with grounding documents methods face two main challenges: (1) they struggle to understand complex multihop relations in long documents, often overlooking subtle factual errors; (2) most specialized methods rely on pairwise comparisons, requiring multiple model calls, leading to high resource and computational costs. To address these challenges, we propose GraphCheck, a fact-checking framework that uses extracted knowledge graphs to enhance text representation. Graph Neural Networks further process these graphs as a soft prompt, enabling LLMs to incorporate structured knowledge more effectively. Enhanced with graph-based reasoning, GraphCheck captures multihop reasoning chains that are often overlooked by existing methods, enabling precise and efficient fact-checking in a single inference call. Experimental results on seven benchmarks spanning both general and medical domains demonstrate up to a 7.1% overall improvement over baseline models. Notably, GraphCheck outperforms existing specialized fact-checkers and achieves comparable performance with state-of-the-art LLMs, such as DeepSeek-V3 and OpenAI-o1, with significantly fewer parameters.

Xiangjue Dong, Cong Wang, Maria Teleki et al.

Persona-assigned Large Language Models (LLMs) can adopt diverse roles, enabling personalized and context-aware reasoning. However, even minor demographic perturbations in personas, such as simple pronoun changes, can alter reasoning trajectories, leading to divergent sets of correct answers. Instead of treating these variations as biases to be mitigated, we explore their potential as a constructive resource to improve reasoning robustness. We propose CHOIR (Collaborative Harmonization fOr Inference Robustness), a test-time framework that harmonizes multiple persona-conditioned reasoning signals into a unified prediction. CHOIR orchestrates a collaborative decoding process among counterfactual personas, dynamically balancing agreement and divergence in their reasoning paths. Experiments on various reasoning benchmarks demonstrate that CHOIR consistently enhances performance across demographics, model architectures, scales, and tasks - without additional training. Improvements reach up to 26.4% for individual demographic groups and 19.2% on average across five demographics. It remains effective even when base personas are suboptimal. By reframing persona variation as a constructive signal, CHOIR provides a scalable and generalizable approach to more reliable LLM reasoning.

Mahsa Valizadeh, Xiangjue Dong, Rui Tuo et al.

Large Language Models (LLMs) excel at capturing latent semantics and contextual relationships across diverse modalities. However, in modeling user behavior from sequential interaction data, performance often suffers when such semantic context is limited or absent. We introduce LaMAR, a LLM-driven semantic enrichment framework designed to enrich such sequences automatically. LaMAR leverages LLMs in a few-shot setting to generate auxiliary contextual signals by inferring latent semantic aspects of a user's intent and item relationships from existing metadata. These generated signals, such as inferred usage scenarios, item intents, or thematic summaries, augment the original sequences with greater contextual depth. We demonstrate the utility of this generated resource by integrating it into benchmark sequential modeling tasks, where it consistently improves performance. Further analysis shows that LLM-generated signals exhibit high semantic novelty and diversity, enhancing the representational capacity of the downstream models. This work represents a new data-centric paradigm where LLMs serve as intelligent context generators, contributing a new method for the semi-automatic creation of training data and language resources.

Maria Teleki, Sai Janjur, Haoran Liu et al.

Evaluating disfluency removal in speech requires more than aggregate token-level scores. Traditional word-based metrics such as precision, recall, and F1 (E-Scores) capture overall performance but cannot reveal why models succeed or fail. We introduce Z-Scores, a span-level linguistically-grounded evaluation metric that categorizes system behavior across distinct disfluency types (EDITED, INTJ, PRN). Our deterministic alignment module enables robust mapping between generated text and disfluent transcripts, allowing Z-Scores to expose systematic weaknesses that word-level metrics obscure. By providing category-specific diagnostics, Z-Scores enable researchers to identify model failure modes and design targeted interventions -- such as tailored prompts or data augmentation -- yielding measurable performance improvements. A case study with LLMs shows that Z-Scores uncover challenges with INTJ and PRN disfluencies hidden in aggregate F1, directly informing model refinement strategies.

Maria Teleki, Xiangjue Dong, Haoran Liu et al.

Masculine defaults are widely recognized as a significant type of gender bias, but they are often unseen as they are under-researched. Masculine defaults involve three key parts: (i) the cultural context, (ii) the masculine characteristics or behaviors, and (iii) the reward for, or simply acceptance of, those masculine characteristics or behaviors. In this work, we study discourse-based masculine defaults, and propose a twofold framework for (i) the large-scale discovery and analysis of gendered discourse words in spoken content via our Gendered Discourse Correlation Framework (GDCF); and (ii) the measurement of the gender bias associated with these gendered discourse words in LLMs via our Discourse Word-Embedding Association Test (D-WEAT). We focus our study on podcasts, a popular and growing form of social media, analyzing 15,117 podcast episodes. We analyze correlations between gender and discourse words -- discovered via LDA and BERTopic -- to automatically form gendered discourse word lists. We then study the prevalence of these gendered discourse words in domain-specific contexts, and find that gendered discourse-based masculine defaults exist in the domains of business, technology/politics, and video games. Next, we study the representation of these gendered discourse words from a state-of-the-art LLM embedding model from OpenAI, and find that the masculine discourse words have a more stable and robust representation than the feminine discourse words, which may result in better system performance on downstream tasks for men. Hence, men are rewarded for their discourse patterns with better system performance by one of the state-of-the-art language models -- and this embedding disparity is a representational harm and a masculine default.

Haoran Liu, Youzhi Luo, Tianxiao Li et al.

We consider the conditional generation of 3D drug-like molecules with \textit{explicit control} over molecular properties such as drug-like properties (e.g., Quantitative Estimate of Druglikeness or Synthetic Accessibility score) and effectively binding to specific protein sites. To tackle this problem, we propose an E(3)-equivariant Wasserstein autoencoder and factorize the latent space of our generative model into two disentangled aspects: molecular properties and the remaining structural context of 3D molecules. Our model ensures explicit control over these molecular attributes while maintaining equivariance of coordinate representation and invariance of data likelihood. Furthermore, we introduce a novel alignment-based coordinate loss to adapt equivariant networks for auto-regressive de-novo 3D molecule generation from scratch. Extensive experiments validate our model's effectiveness on property-guided and context-guided molecule generation, both for de-novo 3D molecule design and structure-based drug discovery against protein targets.

Jianling Wang, Kaize Ding, Ziwei Zhu et al.

Session-based recommender systems aim to improve recommendations in short-term sessions that can be found across many platforms. A critical challenge is to accurately model user intent with only limited evidence in these short sessions. For example, is a flower bouquet being viewed meant as part of a wedding purchase or for home decoration? Such different perspectives greatly impact what should be recommended next. Hence, this paper proposes a novel session-based recommendation system empowered by hypergraph attention networks. Three unique properties of the proposed approach are: (i) it constructs a hypergraph for each session to model the item correlations defined by various contextual windows in the session simultaneously, to uncover item meanings; (ii) it is equipped with hypergraph attention layers to generate item embeddings by flexibly aggregating the contextual information from correlated items in the session; and (iii) it aggregates the dynamic item representations for each session to infer the general purpose and current need, which is decoded to infer the next interesting item in the session. Through experiments on three benchmark datasets, we find the proposed model is effective in generating informative dynamic item embeddings and providing more accurate recommendations compared to the state-of-the-art.

Kaize Ding, Jianling Wang, James Caverlee et al.

Inspired by the extensive success of deep learning, graph neural networks (GNNs) have been proposed to learn expressive node representations and demonstrated promising performance in various graph learning tasks. However, existing endeavors predominately focus on the conventional semi-supervised setting where relatively abundant gold-labeled nodes are provided. While it is often impractical due to the fact that data labeling is unbearably laborious and requires intensive domain knowledge, especially when considering the heterogeneity of graph-structured data. Under the few-shot semi-supervised setting, the performance of most of the existing GNNs is inevitably undermined by the overfitting and oversmoothing issues, largely owing to the shortage of labeled data. In this paper, we propose a decoupled network architecture equipped with a novel meta-learning algorithm to solve this problem. In essence, our framework Meta-PN infers high-quality pseudo labels on unlabeled nodes via a meta-learned label propagation strategy, which effectively augments the scarce labeled data while enabling large receptive fields during training. Extensive experiments demonstrate that our approach offers easy and substantial performance gains compared to existing techniques on various benchmark datasets.

Jianling Wang, Kaize Ding, James Caverlee

A fundamental challenge for sequential recommenders is to capture the sequential patterns of users toward modeling how users transit among items. In many practical scenarios, however, there are a great number of cold-start users with only minimal logged interactions. As a result, existing sequential recommendation models will lose their predictive power due to the difficulties in learning sequential patterns over users with only limited interactions. In this work, we aim to improve sequential recommendation for cold-start users with a novel framework named MetaTL, which learns to model the transition patterns of users through meta-learning. Specifically, the proposed MetaTL: (i) formulates sequential recommendation for cold-start users as a few-shot learning problem; (ii) extracts the dynamic transition patterns among users with a translation-based architecture; and (iii) adopts meta transitional learning to enable fast learning for cold-start users with only limited interactions, leading to accurate inference of sequential interactions.

Monika Daryani, James Caverlee

Fake reviews and review manipulation are growing problems on online marketplaces globally. Review Hijacking is a new review manipulation tactic in which unethical sellers "hijack" an existing product page (usually one with many positive reviews), then update the product details like title, photo, and description with those of an entirely different product. With the earlier reviews still attached, the new item appears well-reviewed. However, there are no public datasets of review hijacking and little is known in the literature about this tactic. Hence, this paper proposes a three-part study: (i) we propose a framework to generate synthetically labeled data for review hijacking by swapping products and reviews; (ii) then, we evaluate the potential of both a Twin LSTM network and BERT sequence pair classifier to distinguish legitimate reviews from hijacked ones using this data; and (iii) we then deploy the best performing model on a collection of 31K products (with 6.5 M reviews) in the original data, where we find 100s of previously unknown examples of review hijacking.

Kaize Ding, Jianling Wang, Jundong Li et al.

Graphs are widely used to model the relational structure of data, and the research of graph machine learning (ML) has a wide spectrum of applications ranging from drug design in molecular graphs to friendship recommendation in social networks. Prevailing approaches for graph ML typically require abundant labeled instances in achieving satisfactory results, which is commonly infeasible in real-world scenarios since labeled data for newly emerged concepts (e.g., new categorizations of nodes) on graphs is limited. Though meta-learning has been applied to different few-shot graph learning problems, most existing efforts predominately assume that all the data from those seen classes is gold-labeled, while those methods may lose their efficacy when the seen data is weakly-labeled with severe label noise. As such, we aim to investigate a novel problem of weakly-supervised graph meta-learning for improving the model robustness in terms of knowledge transfer. To achieve this goal, we propose a new graph meta-learning framework -- Graph Hallucination Networks (Meta-GHN) in this paper. Based on a new robustness-enhanced episodic training, Meta-GHN is meta-learned to hallucinate clean node representations from weakly-labeled data and extracts highly transferable meta-knowledge, which enables the model to quickly adapt to unseen tasks with few labeled instances. Extensive experiments demonstrate the superiority of Meta-GHN over existing graph meta-learning studies on the task of weakly-supervised few-shot node classification.

Ziwei Zhu, Jianling Wang, James Caverlee

Recommendation algorithms typically build models based on historical user-item interactions (e.g., clicks, likes, or ratings) to provide a personalized ranked list of items. These interactions are often distributed unevenly over different groups of items due to varying user preferences. However, we show that recommendation algorithms can inherit or even amplify this imbalanced distribution, leading to unfair recommendations to item groups. Concretely, we formalize the concepts of ranking-based statistical parity and equal opportunity as two measures of fairness in personalized ranking recommendation for item groups. Then, we empirically show that one of the most widely adopted algorithms -- Bayesian Personalized Ranking -- produces unfair recommendations, which motivates our effort to propose the novel fairness-aware personalized ranking model. The debiased model is able to improve the two proposed fairness metrics while preserving recommendation performance. Experiments on three public datasets show strong fairness improvement of the proposed model versus state-of-the-art alternatives. This is paper is an extended and reorganized version of our SIGIR 2020~\cite{zhu2020measuring} paper. In this paper, we re-frame the studied problem as `item recommendation fairness' in personalized ranking recommendation systems, and provide more details about the training process of the proposed model and details of experiment setup.

Yun He, Ziwei Zhu, Yin Zhang et al.

Knowledge of a disease includes information of various aspects of the disease, such as signs and symptoms, diagnosis and treatment. This disease knowledge is critical for many health-related and biomedical tasks, including consumer health question answering, medical language inference and disease name recognition. While pre-trained language models like BERT have shown success in capturing syntactic, semantic, and world knowledge from text, we find they can be further complemented by specific information like knowledge of symptoms, diagnoses, treatments, and other disease aspects. Hence, we integrate BERT with disease knowledge for improving these important tasks. Specifically, we propose a new disease knowledge infusion training procedure and evaluate it on a suite of BERT models including BERT, BioBERT, SciBERT, ClinicalBERT, BlueBERT, and ALBERT. Experiments over the three tasks show that these models can be enhanced in nearly all cases, demonstrating the viability of disease knowledge infusion. For example, accuracy of BioBERT on consumer health question answering is improved from 68.29% to 72.09%, while new SOTA results are observed in two datasets. We make our data and code freely available.

Yun He, Zhuoer Wang, Yin Zhang et al.

We present a new benchmark dataset called PARADE for paraphrase identification that requires specialized domain knowledge. PARADE contains paraphrases that overlap very little at the lexical and syntactic level but are semantically equivalent based on computer science domain knowledge, as well as non-paraphrases that overlap greatly at the lexical and syntactic level but are not semantically equivalent based on this domain knowledge. Experiments show that both state-of-the-art neural models and non-expert human annotators have poor performance on PARADE. For example, BERT after fine-tuning achieves an F1 score of 0.709, which is much lower than its performance on other paraphrase identification datasets. PARADE can serve as a resource for researchers interested in testing models that incorporate domain knowledge. We make our data and code freely available.

Habeeb Hooshmand, James Caverlee

A large portion of the car-buying experience in the United States involves interactions at a car dealership. At the dealership, the car-buyer relays their needs to a sales representative. However, most car-buyers are only have an abstract description of the vehicle they need. Therefore, they are only able to describe their ideal car in "car-speak". Car-speak is abstract language that pertains to a car's physical attributes. In this paper, we define car-speak. We also aim to curate a reasonable data set of car-speak language. Finally, we train several classifiers in order to classify car-speak.