论文检索

Dayeon Ki, Yu Hou, Rachel Rudinger, Hal Daumé Iii, Marine Carpuat, Fumeng Yang

Neologisms and emerging slang are central to daily conversation, yet challenging for non-native speakers (NNS) to interpret and use appropriately in cross-cultural communication with native speakers (NS). NNS increasingly make use of Artificial Intelligence (AI) tools to learn these words. We study the utility of such tools in mediating an informal communication scenario through a human-subjects study (N=234): NNS participants learn English neologisms with AI support, write messages using the learned word to an NS friend, and judge contextual appropriateness of the neologism in two provided writing samples. Using both NS evaluator-rated communicative competence of NNS-produced writing and NNS’ contextual appropriateness judgments, we compare three AI-based support conditions: AI Definition, AI Rewrite into simpler English, AI Explanation of meaning and usage, and Non-AI Dictionary for comparison. We show that AI Explanation yields the largest gains over no support in NS-rated competence, while contextual appropriateness judgments show indifference across support. NNS participants’ self-reported perceptions tend to overestimate NS ratings, revealing a mismatch between perceived and actual competence. We further observe a significant gap between NNS- and NS-produced writing, highlighting the limitations of current AI tools and informing design for future tools.

Yunchong Huang, Gianni Barlacchi, Sandro Pezzelle

Large language models (LLMs) perform well on well-posed questions, yet standard question-answering (QA) benchmarks remain far from solved. We argue that this gap is partly due to underspecified questions, that are queries whose interpretation cannot be uniquely determined without additional context. We introduce an LLM-based classifier to identify underspecified questions and apply it to several widely used QA datasets, finding that 16% to over 60% of benchmark questions are underspecified and that LLMs perform significantly worse on them. To isolate the effect of underspecification, we conduct a controlled rewriting experiment that serves as an upper-bound analysis, rewriting underspecified questions into fully specified variants while holding gold answers fixed. QA performance consistently improves under this setting, indicating that many apparent QA failures stem from question underspecification rather than model limitations. Our findings highlight underspecification as an important confound in QA evaluation and motivate greater attention to question clarity in benchmark design.

Yizheng Zhao

Recent theory suggests Transformers are inherently succinct, capable of representing recursive algorithms like binary counting over exponential state spaces using constant-size circuits, unlike the exponential bottleneck of RNNs. However, it remains unclear under what conditions gradient-trained Transformers converge to these predicted succinct circuits, or whether they settle for heuristics. We bridge this gap by rigorously testing the Succinctness Hypothesis via mechanistic interpretability on the LargeCounter task. We report a striking dichotomy: shallow Transformers (d=64) generalize perfectly, whereas massive LSTM baselines (d=2048) fail completely (<6\% accuracy), empirically validating the succinctness gap. This dichotomy extends to modern state-space models: Mamba and Mamba-2 fail even more catastrophically (<1.1\%), confirming the hierarchy Transformer \gg LSTM > SSM predicted by formal complexity results. We show this capability is acquired via a grokking phase transition driven by a weight-norm "complexity collapse". Mechanistic analysis reveals the learned circuit aligns precisely with Boolean RASP theory: attention heads utilize Rotary Positional Embeddings (RoPE) for "Same-Bit Lookup", while MLPs act as exact XOR/AND logic gates. Furthermore, we detect analogous "Arithmetic Heads" in pre-trained LLMs (Pythia), suggesting that succinctness is a representational inductive bias that, when activated by sufficient regularization, governs how models learn algorithmic reasoning.

Ahson Saiyed, Sabrina Sadiekh, Chirag Agarwal

Large Language Models (LLMs) remain vulnerable to optimization-based jailbreak attacks that exploit internal gradient structure. While Sparse Autoencoders (SAEs) are widely used for interpretability, their robustness implications remain underexplored. We present a study of integrating pretrained SAEs into transformer residual streams at inference time, without modifying model weights or blocking gradients. Across four model families (Gemma, LLaMA, Mistral, Qwen) and two strong white-box attacks (GCG, BEAST) plus three black-box benchmarks, SAE-augmented models achieve up to a 5x reduction in jailbreak success rate relative to the undefended baseline and reduce cross-model attack transferability. Parametric ablations reveal (i) a monotonic dose-response relationship between L0 sparsity and attack success rate, and (ii) a layer-dependent defense-utility tradeoff, where intermediate layers balance robustness and clean performance. These findings are consistent with a representational bottleneck hypothesis: sparse projection reshapes the optimization geometry exploited by jailbreak attacks.

Ruixing Zhang, Zihan Liu, Leilei Sun, Tongyu Zhu, Weifeng Lv

Geo-localization aims to infer the geographic origin of a given signal. In computer vision, geo-localization has served as a demanding benchmark for compositional reasoning and is relevant to public safety. In contrast, progress on audio geo-localization has been constrained by the lack of high-quality audio-location pairs. To address this gap, we introduce AGL1K, the first audio geo-localization benchmark for audio language models (ALMs), spanning 72 countries and territories. To extract reliably localizable samples from a crowd-sourced platform, we propose the Audio Localizability metric that quantifies the informativeness of each recording, yielding 1,444 curated audio clips. Evaluations on 16 ALMs show that ALMs have emerged with audio geo-localization capability. We find that closed-source models substantially outperform open-source models, and that linguistic clues often dominate as a scaffold for prediction. We further analyze ALMs’ reasoning traces, regional bias, error causes, and the interpretability of the localizability metric. Overall, AGL1K establishes a benchmark for audio geo-localization and may advance ALMs with better geospatial reasoning capability.

Nisrine Rair, Alban Goupil, Valeriu Vrabie, Emmanuel Chochoy

Subjective NLP datasets typically aggregate annotator judgments into a single gold label, making it difficult to diagnose whether disagreement reflects unclear criteria, collapsed distinctions, or legitimate plurality. We propose a schema-level diagnostic for auditing expert-designed annotation schemas prior to gold-label commitment, using only multi-annotator criterion judgments. The diagnostic separates two failure modes: unstable criteria with hard-to-operationalize boundaries, and systematic overlap that blurs the boundaries between mutually exclusive categories. Applied to persuasive value extraction in commercial documents, we find that disagreement is not diffuse: instability concentrates in a few criteria, while nearly half of covered sentences activate multiple categories. These signals align with where domain experts disagree, yielding an evidence-based audit for tightening guidelines, revising category structure, or reconsidering the annotation paradigm.

Hoonsang Yoon, Takyoung Kim, Wonkee Lee, Ilmin Cho, Dilek Hakkani-Tür, Stanley Jungkyu Choi

Traditional topic modeling treats each document as a single, coherent unit of topic, which can cause topic contamination when documents cover multiple topics. This becomes especially problematic when stakeholders are interested in identifying documents that focus on a specific topic. We introduce segment-based topic allocation, a novel paradigm that redefines topic assignment at the level of segments, coherent textual spans conveying distinct topical content. This granularity improves topic purity, interpretability, and applicability to multi-theme corpora such as reviews or survey responses. To support this paradigm, we construct SemEval-STM, a benchmark derived from aspect-based sentiment datasets, where segments are automatically extracted using large language models (LLMs) and post-processed with human supervision. We further propose the segment intrusion task (SIT), a novel evaluation method extending word intrusion to the span level, enabling human-centric assessment of topical coherence. Empirical results across diverse metrics and models demonstrate that SBTA significantly outperforms traditional document-based methods in clustering and interpretability. Our framework provides a practical and scalable solution for fine-grained topic analysis in heterogeneous text corpora.

Maisha Maliha, Dean F. Hougen

Text-to-image diffusion models achieve remarkable generation quality, yet their internal mechanisms for grounding prompt semantics into visual structure remain poorly understood. We present a novel mechanistic interpretability framework for Stable Diffusion that probes how individual prompt tokens are represented and utilized during the denoising process. Given a prompt, we record cross-attention activations throughout UNet denoising and convert them into token-level spatial grounding maps that indicate where each token contributes signal during image synthesis. To establish causal faithfulness, we perform controlled prompt interventions by removing a single word at a time while keeping the sampling seed fixed, producing counterfactual generations. To quantify mechanistic sensitivity, we introduce a head-resolved spike score based on divergence between per-head token contribution distributions before and after intervention, enabling module-wise and head-wise attribution of semantic changes. Experiments on compositional prompts and challenging relational descriptions reveal systematic patterns of token grounding, semantic drift, and head specialization across denoising timesteps. Our results provide a practical and reproducible toolkit for analyzing how diffusion models encode and apply semantic information, supporting deeper transparency in text-to-image generation.

Zhaolin Li, Jan Niehues

Automatic speech recognition (ASR) still covers only a small fraction of the world’s languages, mainly due to supervised data scarcity. In-context learning (ICL) with large language models (LLMs) addresses this problem, but prior work largely focuses on high-resource languages covered during training and text-only settings. This paper investigates whether speech LLMs can learn unseen languages with multimodal ICL (MICL), and how this learning can be used to improve ASR. We conduct experiments with two speech LLMs, Phi-4 and Qwen3-Omni, on three diverse endangered languages. Firstly, we find that MICL is effective for unseen languages, leveraging both speech and text modalities. We further show that cross-lingual transfer learning improves MICL efficiency on target languages without training on them. Moreover, we analyze attention patterns to interpret MICL mechanisms, and we observe layer-dependent preferences between audio and text context, with an overall bias towards text. Finally, we show that prompt-based ASR with speech LLMs performs poorly on unseen languages, motivating a simple ASR system that combines a stronger acoustic model with a speech LLM via MICL-based selection of acoustic hypotheses. Results show that MICL consistently improves ASR performance, and that cross-lingual transfer learning matches or outperforms corpus-trained language models without using target-language data. Our code is publicly available

Yibo Yan, Shen Wang, Jiahao Huo, Jingheng Ye, Zhendong Chu, Xuming Hu, Philip S. Yu, Carla P Gomes, Bart Selman, Qingsong Wen

Scientific reasoning, the process through which humans apply logic, evidence, and critical thinking to explore and interpret scientific phenomena, is essential in advancing knowledge reasoning across diverse fields. However, despite significant progress, current scientific reasoning models still struggle with generalization across domains and often fall short of multimodal perception. Multimodal Large Language Models (MLLMs), which integrate text, images, and other modalities, present an exciting opportunity to overcome these limitations and enhance scientific reasoning. Therefore, **this position paper argues that MLLMs can significantly advance scientific reasoning across disciplines such as mathematics, physics, chemistry, and biology**. We highlight the current state of MLLM applications in scientific reasoning, noting their ability to integrate and reason over diverse data types. However, challenges such as multimodal alignment, data diversity, and reasoning depth remain obstacles to achieving their full potential. To address these challenges, we propose actionable suggestions in the near future. Overall, our work offers a novel perspective on MLLM integration with scientific reasoning, providing the LLM community with valuable insights for achieving Artificial General Intelligence (AGI).

Xiangning Yu, Yuwei Guo, Yuqi Hou, Xiao Xue, Qun Ma

LLM-empowered agent simulations are increasingly used to study social emergence, yet the micro-to-macro causal mechanisms behind macro outcomes often remain unclear. This is challenging because emergence arises from intertwined agent interactions and meso-level feedback and nonlinearity, making generative mechanisms hard to disentangle. To this end, we introduce CAMO, an automated Causal discovery framework from Micro behaviors to Macro Emergence in LLM agent simulations. CAMO converts mechanistic hypotheses into computable factors grounded in simulation records and learns a compact causal representation centered on an emergent target . CAMO outputs a computable Markov boundary and a minimal upstream explanatory subgraph, yielding interpretable causal chains and actionable intervention levers. It also uses simulator-internal counterfactual probing to orient ambiguous edges and revise hypotheses when evidence contradicts the current view. Experiments across four emergent settings demonstrate the promise of CAMO.[The code is available at an anonymous link: <https://anonymous.4open.science/r/CAMO-0E6C/>.]

Ine Gevers, Walter Daelemans

Large language models (LLMs) have achieved striking successes on many benchmarks, yet recent studies continue to expose fundamental weaknesses. In this paper, we introduce Concept, a simple word-guessing board game, as a benchmark for probing abductive reasoning. Our results show that this game, easily solved by humans (with a success rate of over 90%), is still very challenging for state-of-the-art LLMs (no model exceeds 40% success rate). Specifically, we observe that LLMs struggle with interpreting other players’ strategic intents, and with correcting initial hypotheses given sequential information updates. In addition, we extend the evaluation across multiple languages, and find that the LLM performance drops further in lower-resource languages (Dutch, French, and Spanish) compared to English.

Loris Schoenegger, Benjamin Roth

Training data influence estimation methods quantify the contribution of training documents to a model’s output, making them a promising source of information for example-based explanations.As humans cannot interpret thousands of documents, only a small subset of the training data can be presented as an explanation.Although the choice of which documents to include directly affects explanation quality, previous evaluations of such systems have largely ignored any selection strategies.To address this, we propose a novel *selection relevance score*, a retraining-free metric that quantifies how useful a set of examples is for explaining a model’s output.We validate this score through fine-tuning experiments, confirming that it can predict whether a set of examples supports or undermines the model’s predictions.Using this metric, we further show that common selection strategies often underperform random selection. Motivated by this finding, we propose a strategy that balances influence and representativeness, enabling better use of selection budgets than naively selecting the highest-ranking examples.

Shuoxin Wang, Chang Liu, Gowen Loo, Lifan Zheng, Kaiwen Wei, Huanqian Yan, Xinyi Zeng, Jingyuan Zhang, Yu Tian

Large Language Model (LLM)-based mobile agents have made significant performance advancements. However, these agents often follow explicit user instructions while overlooking personalized needs, leading to significant limitations for real users, particularly without personalized context: (1) inability to interpret ambiguous instructions, (2) lack of learning from user interaction history, and (3) failure to handle personalized instructions. To alleviate the above challenges, we propose Me-Agent, a learnable and memorable personalized mobile agent. Specifically, Me-Agent incorporates a two-level user habit learning approach. At the prompt level, we design a user preference learning strategy enhanced with a Personal Reward Model to improve personalization performance. At the memory level, we design a Hierarchical Preference Memory, which stores users’ long-term memory and app-specific memory in different level memory. To validate the personalization capabilities of mobile agents, we introduce User FingerTip, a new benchmark featuring numerous ambiguous instructions for daily life. Extensive experiments on User FingerTip and general benchmarks demonstrate that Me-Agent achieves state-of-the-art performance in personalization while maintaining competitive instruction execution performance.

Xuanle Zhao, Xinyuan Cai, Xiang Cheng, Xiuyi Chen, Bo XU

While Vision-Language Models (VLMs) have demonstrated significant potential in chemical visual understanding, current models are predominantly optimized for direct visual question-answering tasks. This paradigm often results in "black-box" systems that fail to utilize the inherent capability of Large Language Models (LLMs) to infer underlying reaction mechanisms. In this work, we introduce ChemVLR, a chemical VLM designed to prioritize reasoning within the perception process. Unlike conventional chemical VLMs, ChemVLR analyzes visual inputs in a fine-grained manner by explicitly identifying granular chemical descriptors, such as functional groups, prior to generating answers. This approach ensures the production of explicit and interpretable reasoning paths for complex visual chemical problems. To facilitate this methodology, we implement a cross-modality reverse-engineering strategy combined with a rigorous filtering pipeline to curate a large-scale reasoning and caption dataset, comprising 760k high-quality samples across molecular and reaction tasks. Furthermore, we adopt a three-stage training framework that systemically builds model perception and reasoning capacity. Experiments demonstrate that ChemVLR achieves state-of-the-art (SOTA) performance, surpassing both leading proprietary models and domain-specific open-source baselines. We also provide comprehensive ablation studies to validate our training strategy and data generation designs.

Chau Minh Pham, Zichao Wang, Puneet Mathur, Alexa Siu, Akriti Jain, Aparna Garimella, Ananya B. Sai, Nedim Lipka, Mohit Iyyer, Varun Manjunatha

Large language models are increasingly used to draft long-form multimodal documents, but their end-to-end performance on professional report generation remains systematically understudied. We introduce AnalystBench, a continually extensible benchmark of 20 real-world report generation tasks grounded in multimodal document collections, where models must process millions of input tokens to produce long-form professional reports. Using expert-validated quality checklists and groundedness evaluation, we evaluate LLMs and coding agents and find that the best model, GPT-5.1, scores highly on executive summarization tasks (exceeding 90% on quality checklists) but degrades substantially on tasks requiring long-horizon synthesis over large inputs (dropping to 25-40%). Agent-based generation substantially benefits strong closed-source models such as GPT-5.1, with checklist scores improving by 20.24 percentage points and visual coverage by 37.41 points over vanilla generation, but offers little or negative gains for open-source models like DeepSeek-R1 (-3.02 points). Expert reviewers note that while generated reports are grounded and clearly separate factual description from interpretation, they often fall short in actionability, clarity, and quantitative precision, which highlights the gap between system performance and real-world professional needs.

Yifei Yao, Hanrong Zhang, Mengnan Du

Understanding the internal representations of large language models (LLMs) remains a central challenge for interpretability research. Sparse autoencoders (SAEs) offer a promising solution by decomposing activations into interpretable features, but existing approaches rely on fixed sparsity constraints that fail to account for input complexity. We propose AdaptiveK SAE (Adaptive Top K Sparse Autoencoders), a novel framework that dynamically adjusts sparsity levels based on the semantic complexity of each input. Leveraging linear probes, we demonstrate that context complexity is linearly encoded in LLM representations, and we use this signal to guide feature allocation during training. Experiments across ten language models demonstrate that this complexity-driven adaptation outperforms fixed-sparsity approaches on reconstruction fidelity, explained variance, cosine similarity and interpretability metrics while eliminating the burden of extensive hyperparameter tuning. Our code is available at: https://github.com/hiyukie/adaptiveK.

Jiatong Li, Hailong Cao, Yang Liu

Large Language Models (LLMs) demonstrate strong performance in multilingual tasks, yet the process of constructing predictions in the target language remains under-explored. In this work, we introduce the FFN Lens, a novel interpretability method focusing on the Transformer’s core computational module, the Feed-Forward Network (FFN). By directly leveraging model parameters, the FFN Lens identifies both the critical units responsible for constructing specific information and the input features that drive them, which is essential for understanding Large Language Models. Applying FFN Lens to multilingual tasks, we demonstrate the prediction construction process and reveal the distinct division of labor across model layers. We identify a three-stage functional pipeline for constructing multilingual predictions: Latent Translation, Semantic Mapping, and Self Emphasis. We further introduce subspace analysis to validate this three-stage mechanism from a complementary perspective, and leverage these mechanistic insights to propose a training-free uncertainty estimation method.

Zhou Ziheng, Jiakun Ding, Zhaowei Zhang, Ruosen Gao, Ying Nian Wu, Demetri Terzopoulos, Yipeng Kang, Fangwei Zhong, Junqi Wang

Principle-based alignment often lacks context sensitivity and completeness. Grounded in Theory of Mind, we propose role conditioning as a compact alternative: social roles (e.g., mother, judge) implicitly encode both values and the cognitive schemas required to apply them. We introduce a training-free pipeline featuring a role-conditioned generator and iterative role-based critics for refinement. Across five model families, our approach consistently outperforms principle-based, Chain-of-Thought (CoT) and other baselines across benchmarks. Notably, it reduces unsafe outputs on the WildJailbreak benchmark from 81.4% to 3.6% with DeepSeek-V3. Not only for common safety benchmarks, it consistently applies for agentic safety tasks. These results establish role assignment as a powerful, interpretable paradigm for AI alignment and LLM-as-a-Judge construction.

Jiazheng Zhang, Ziche Fu, Zhiheng Xi, Wenqing Jing, Mingxu Chai, Wei He, Guoqiang Zhang, Chenghao Fan, Chenxin An, Wenxiang Chen 等

Verifiers have been demonstrated to enhance LLM reasoning via test-time scaling (TTS). Yet, they face significant challenges in complex domains. Error propagation from incorrect intermediate reasoning can lead to false positives for seemingly plausible solutions, while lacking external grounding makes verifiers unreliable on computation or knowledge-intensive tasks. To address these challenges, we propose Agentic Verifier, a framework that transforms reward modeling into a multi-turn, tool-augmented deliberative process. We introduce complementary forward and backward agents: one traces solutions from premises to conclusions, while the other re-checks conclusions against their underlying premises. This bidirectional process enables a comprehensive, reliable, and interpretable assessment of solutions. To facilitate practical deployment, we propose AgentV-RL. Through proactive exploration and reinforcement learning, the verifier autonomously interleaves tool-use with internal reasoning. Extensive experiments show that Agentic Verifier yields consistent performance gains under both parallel and sequential TTS. Notably, our 4B variant surpasses state-of-the-art ORMs by 25.2%, positioning it as a promising paradigm for agentic reward modeling.