论文检索

Applications · Language, Speech and Dialog

Yixuan Weng, Minjun Zhu, Qiujie Xie, QiYao Sun, Zhen Lin, Sifan Liu, Yue Zhang

While previous AI Scientist systems can generate novel findings, they often lack the focus to produce scientifically valuable contributions that address pressing human-defined challenges. We introduce DeepScientist, a system designed to overcome this by conducting goal-oriented, fully autonomous scientific discovery over month-long timelines. It formalizes discovery as a Bayesian Optimization problem, using a cumulative Findings Memory to intelligently balance the exploitation of promising avenues with the exploration of novel hypotheses. Consuming over 20,000 GPU hours, the system generated about 5,000 unique ideas and experimentally validated approximately 1100, ultimately surpassing human-designed 2025 state-of-the-art (SOTA) methods on three frontier AI tasks by 183.7\%, 1.9\%, and 7.9\%. Crucially, this was achieved by autonomously redesigning core methodologies, not merely recombining existing techniques. In a striking demonstration, the system achieved progress on AI text detection in just two weeks that is comparable to three years of cumulative human research. This work provides the first large-scale evidence of an AI achieving discoveries that progressively surpass human SOTA on scientific tasks, producing valuable findings that genuinely push the frontier forward.

Applications · Language, Speech and Dialog

Dingzirui Wang, Xuanliang Zhang, Keyan Xu, Qingfu Zhu, Wanxiang Che, Yang Deng

Existing research indicates that the output of **Chain-of-Thought (CoT)** is significantly affected by input perturbations. Although many methods aim to mitigate such impact by optimizing prompts, a theoretical explanation of how these perturbations influence CoT outputs remains an open area of research. This gap limits our in-depth understanding of how input perturbations propagate during the reasoning process and hinders further improvements in prompt optimization methods. Therefore, in this paper, we theoretically analyze the effect of input perturbations on the fluctuation of CoT outputs. We first derive an upper bound for input perturbations under the condition that the output fluctuation is within an acceptable range, and we prove that: - *i)* This upper bound is **positively correlated** with the number of reasoning steps in the CoT; - *ii)* Even an infinitely long reasoning process **cannot eliminate** the impact of input perturbations. We then apply these conclusions to the **Linear Self-Attention (LSA)** model, which can be viewed as a simplified version of Transformer. For the LSA model, we prove that the upper bound for input perturbation is **negatively correlated** with the norms of the input embedding and hidden state vectors. To validate this theoretical analysis, we conduct experiments on **three mainstream datasets** and **four mainstream models**. The experimental results align with our theoretical analysis, empirically demonstrating the correctness of our findings.

Computer Vision · Image and Video Generation

Jiayi Lyu, Leigang Qu, Wenjing Zhang, Hanyu Jiang, Kai Liu, Zhenglin Zhou, Xiaobo Xia, Jian Xue, Tat-Seng Chua

Realistic talking-head video generation is critical for virtual avatars, film production, and interactive systems. Current methods struggle with nuanced emotional expressions due to the lack of fine-grained emotion control. To address this issue, we introduce a novel two-stage method (AUHead) to disentangle fine-grained emotion control, i.e. , Action Units (AUs), from audio and achieve controllable generation. In the first stage, we explore the AU generation abilities of large audio-language models (ALMs), by spatial-temporal AU tokenization and an "emotion-then-AU" chain-of-thought mechanism. It aims to disentangle AUs from raw speech, effectively capturing subtle emotional cues. In the second stage, we propose an AU-driven controllable diffusion model that synthesizes realistic talking-head videos conditioned on AU sequences. Specifically, we first map the AU sequences into the structured 2D facial representation to enhance spatial fidelity, and then model the AU-vision interaction within cross-attention modules. To achieve flexible AU-quality trade-off control, we introduce an AU disentanglement guidance strategy during inference, further refining the emotional expressiveness and identity consistency of the generated videos. Results on benchmark datasets demonstrate that our approach achieves competitive performance in emotional realism, accurate lip synchronization, and visual coherence, significantly surpassing existing techniques. Our implementation is available at https://github.com/laura990501/AUHead_ICLR

Applications · Language, Speech and Dialog

Yingyi Zhang, Junyi Li, Wenlin Zhang, Pengyue Jia, Xianneng Li, Yichao Wang, Derong Xu, Yi Wen, Huifeng Guo, Yong Liu 等

Personalized large language models (LLMs) rely on memory retrieval to incorporate user-specific histories, preferences, and contexts. Existing approaches either overload the LLM by feeding all the user's past memory into the prompt, which is costly and unscalable, or simplify retrieval into a one-shot similarity search, which captures only surface matches. Cognitive science, however, shows that human memory operates through a dual process: Familiarity, offering fast but coarse recognition, and Recollection, enabling deliberate, chain-like reconstruction for deeply recovering episodic content. Current systems lack both the ability to perform recollection retrieval and mechanisms to adaptively switch between the dual retrieval paths, leading to either insufficient recall or the inclusion of noise. To address this, we propose RF-Mem (Recollection–Familiarity Memory Retrieval), a familiarity uncertainty-guided dual-path memory retriever. RF-Mem measures the familiarity signal through the mean score and entropy. High familiarity leads to the direct top-$K$ Familiarity retrieval path, while low familiarity activates the Recollection path. In the Recollection path, the system clusters candidate memories and applies $\alpha$-mix with the query to iteratively expand evidence in embedding space, simulating deliberate contextual reconstruction. This design embeds human-like dual-process recognition into the retriever, avoiding full-context overhead and enabling scalable, adaptive personalization. Experiments across three benchmarks and corpus scales demonstrate that RF-Mem consistently outperforms both one-shot retrieval and full-context reasoning under fixed budget and latency constraints. Our code can be found in the Reproducibility Statement.

Applications · Language, Speech and Dialog

Zeyu Jin, Songtao Zhou, Haoyu Wang, Minghao Tian, Kaifeng Yun, Zhuo Chen, Xiaoyu Qin, Jia Jia

The recent advancement of Artificial Intelligence Generated Content (AIGC) has led to significant strides in modeling human interaction, particularly in the context of multimodal dialogue. While current methods impressively generate realistic dialogue in isolated modalities like speech or vision, challenges remain in controllable Multimodal Dialogue Generation (MDG). This paper focuses on the natural alignment between speech, vision, and text in human interaction, aiming for expressive dialogue generation through multimodal conditional control. To address the insufficient richness and diversity of dialogue expressiveness in existing datasets, we introduce a novel multimodal dialogue annotation pipeline to curate dialogues from movies and TV series with fine-grained annotations in interactional characteristics. The resulting MM-Dia dataset (360+ hours, 54,700 dialogues) facilitates explicitly controlled MDG, specifically through style-controllable dialogue speech synthesis. In parallel, MM-Dia-Bench (309 highly expressive dialogues with visible single-/dual-speaker scenes) serves as a rigorous testbed for implicit cross-modal MDG control, evaluating audio-visual style consistency across modalities. Extensive experiments demonstrate that training on MM-Dia significantly enhances fine-grained controllability, while benchmarks on MM-Dia-Bench reveal limitations in current frameworks to replicate the nuanced expressiveness of human interaction. These findings provides new insights and challenges for multimodal conditional dialogue generation.

Applications · Language, Speech and Dialog

Maojia Song, Tej Deep Pala, Ruiwen Zhou, Weisheng Jin, Amir Zadeh, Chuan Li, Dorien Herremans, Soujanya Poria

Large language models (LLMs) are increasingly deployed in multi-agent systems (MAS) as components of collaborative intelligence, where peer interactions dynamically shape individual decision-making. While prior work has largely focused on conformity bias, we broaden the scope to examine how LLMs build rapport from previous interactions, resist misinformation, and integrate peer input during collaboration, which are key factors for achieving collective intelligence under complex social dynamics. We introduce KAIROS, a benchmark simulating quiz contests with peer agents of varying reliability, offering fine-grained control over conditions such as expert–novice roles, noisy crowds, and adversarial peers. LLMs receive both historical interactions and current peer responses, allowing systematic investigation into how rapport, peer action, and self-confidence influence decisions. To mitigate this vulnerability, we evaluate prompting, supervised fine-tuning, and reinforcement learning using Group Relative Policy Optimization (GRPO) across multiple models. Our results show that model size plays a central role in moderating susceptibility to social influence: larger models exhibit stronger resilience and benefit from prompting-based mitigation, whereas smaller models are more vulnerable. For the latter, carefully configured GRPO training improves both robustness and overall performance. Our code and datasets are available at: https://anonymous.4open.science/r/KAIROS-4F71

Applications · Language, Speech and Dialog

Siru Ouyang, Jun Yan, I-Hung Hsu, Yanfei Chen, Ke Jiang, Zifeng Wang, Rujun Han, Long Le, Samira Daruki, Xiangru Tang 等

With the growing adoption of large language model agents in persistent real-world roles, they naturally encounter continuous streams of tasks. A key limitation, however, is their failure to learn from the accumulated interaction history, forcing them to discard valuable insights and repeat past errors. We propose ReasoningBank, a novel memory framework that distills generalizable reasoning strategies from an agent's self-judged successful and failed experiences. At test time, an agent retrieves relevant memories from ReasoningBank to inform its interaction and then integrates new learnings back, enabling it to become more capable over time. Building on this powerful experience learner, we further introduce memory-aware test-time scaling (MaTTS), which accelerates and diversifies this learning process by scaling up the agent's interaction experience. By allocating more compute to each task, the agent generates abundant, diverse experiences that provide rich contrastive signals for synthesizing higher-quality memory. The better memory in turn guides more effective scaling, establishing a powerful synergy between memory and test-time scaling. Across web browsing and software engineering benchmarks, ReasoningBank consistently outperforms existing memory mechanisms that store raw trajectories or only successful task routines, improving both effectiveness and efficiency; MaTTS further amplifies these gains. These findings establish memory-driven experience scaling as a new scaling dimension, enabling agents to self-evolve with emergent behaviors naturally arise. Our code can be found at https://github.com/google-research/reasoning-bank.

Applications · Language, Speech and Dialog

Dmitrii Korzh, Dmitrii Tarasov, Artyom Iudin, Elvir Karimov, Matvey Skripkin, Nikita Kuzmin, Andrey Kuznetsov, Oleg Rogov, Ivan Oseledets

Conversion of spoken mathematical expressions is a challenging task that involves transcribing speech into a strictly structured symbolic representation while addressing the ambiguity inherent in the pronunciation of equations. Although significant progress has been achieved in automatic speech recognition (ASR) and language models (LM), the problem of converting spoken mathematics into LaTeX remains underexplored. This task directly applies to educational and research domains, such as lecture transcription or note creation. Based on ASR post-correction, prior work requires 2 transcriptions, focuses only on isolated equations, has a limited test set, and provides neither training data nor multilingual coverage. To address these issues, we present the first fully open-source large-scale dataset, comprising over 66,000 human-annotated audio samples of mathematical equations and sentences in English and Russian, drawn from diverse scientific domains. In addition to the ASR post-correction models and few-shot prompting, we apply audio language models, demonstrating comparable character error rate (CER) results on the MathSpeech benchmark (28\% vs. 30\%) for the equations conversion. In contrast, on the proposed S2L-equations benchmark, our models outperform the MathSpeech model by a substantial margin of more than 36 percentage points, even after accounting for LaTeX formatting artifacts (27\% vs. 64\%). We establish the first benchmark for mathematical sentence recognition (S2L-sentences) and achieve an equation CER of 40\%. This work lays the groundwork for future advances in multimodal AI, with a particular focus on mathematical content recognition.

Applications · Language, Speech and Dialog

Guocong Li, Jinjian Zhang, Ping Wang, Dongnan Liu, Tian Liang, Qiuyi Qi, Hao Huang, Siyan Guo, Mutian Bao, Wei Zhou 等

We present Mixture-of-Length (MoL), an approach for Question Answering (QA) with context that aims to improve the balance between reasoning quality and response efficiency. Our method introduces a principled difficulty assessment based on information-theoretic principles and a dual-objective reward mechanism that adaptively modulates response length. In our experiments, MoL exhibits an emergent behavior termed "intelligent brevity": the model tends to produce shorter responses for simpler queries and longer ones for more complex inputs. This property is desirable for human-computer interaction and can reduce inference costs. A post-hoc analysis of internal activations suggests a correlation between this output adaptivity and the effective number of layers that contribute during inference. On multiple QA benchmarks, MoL demonstrates competitive accuracy while substantially reducing tokens compared to baselines, indicating that difficulty-aware length modulation is a promising direction for efficient QA with context.

Applications · Language, Speech and Dialog

Derong Xu, Yi Wen, Pengyue Jia, Yingyi Zhang, Wenlin Zhang, Yichao Wang, Huifeng Guo, Ruiming Tang, Xiangyu Zhao, Enhong Chen 等

Large Language Models (LLMs) have recently been widely adopted in conversational agents. However, the increasingly long interactions between users and agents accumulate extensive dialogue records, making it difficult for LLMs with limited context windows to maintain a coherent long-term dialogue memory and deliver personalized responses. While retrieval-augmented memory systems have emerged to address this issue, existing methods often depend on single-granularity memory segmentation and retrieval. This approach falls short in capturing deep memory connections, leading to partial retrieval of useful information or substantial noise, resulting in suboptimal performance. To tackle these limits, we propose MemGAS, a framework that enhances memory consolidation by constructing multi-granularity association, adaptive selection, and retrieval. MemGAS is based on multi-granularity memory units and employs Gaussian Mixture Models to cluster and associate new memories with historical ones. An entropy-based router adaptively selects optimal granularity by evaluating query relevance distributions and balancing information completeness and noise. Retrieved memories are further refined via LLM-based filtering. Experiments on four long-term memory benchmarks demonstrate that MemGAS outperforms state-of-the-art methods on both question answer and retrieval tasks, achieving superior performance across different query types and top-K settings\footnote{https://github.com/Applied-Machine-Learning-Lab/ICLR2026\_MemGAS}.

Applications · Language, Speech and Dialog

Kai Xiong, Xiao Ding, Yixin Cao, Li Du, jiahao ying, yang zhao, Bing Qin, Ting Liu

Large Language Models (LLMs) have demonstrated impressive generalization ability by learning from extensive unlabeled text. However, they still exhibit reasoning mistakes, which can affect their trustworthiness and reliability. Although users can interact with LLMs and provide diverse and comprehensive queries to expose the flaws of LLMs, obtaining sufficient and effective feedback is demanding. Furthermore, comprehensively evaluating LLMs with limited labeled samples is difficult. These make it a challenge to diagnose and remedy the deficiencies in LLMs through rich label-free user queries. To tackle this challenge and considersing that LLMs' reasoning mistakes often stem from knowledge deficiencies, we propose label-free curricular meaningful learning (LaMer), which first employs relative entropy to diagnose and quantify knowledge deficiencies of LLMs in a label-free setting. Then, LaMer adaptively synthesizes augmentation data based on deficiency severity and progressively remedies them with a curricular remedy strategy. Experiments show that LaMer effectively diagnoses and remedies knowledge deficiencies in LLMs, improving various LLMs across seven out-of-distribution (OOD) reasoning benchmarks, achieving comparable results to baselines with only 40% training data. LaMer even surpasses methods that rely on labeled data for deficiency diagnosis. In application, LaMer offers a diagnostic tool for efficient LLM development.

Applications · Language, Speech and Dialog

Qihao Liu, Luoxin Ye, Wufei Ma, Yu-Cheng Chou, Alan Yuille

Large language models (LLMs) with explicit reasoning capabilities excel at mathematical reasoning yet still commit process errors, such as incorrect calculations, brittle logic, and superficially plausible but invalid steps. In this paper, we introduce Generative Adversarial Reasoner, an on-policy joint training framework designed to enhance reasoning by co-evolving an LLM reasoner and an LLM-based discriminator through adversarial reinforcement learning. A compute-efficient review schedule partitions each reasoning chain into logically complete slices of comparable length, and the discriminator evaluates each slice’s soundness with concise, structured justifications. Learning couples complementary signals: the LLM reasoner is rewarded for logically consistent steps that yield correct answers, while the discriminator earns rewards for correctly detecting errors or distinguishing traces in the reasoning process. This produces dense, well-calibrated, on-policy step-level rewards that supplement sparse exact-match signals, improving credit assignment, increasing sample efficiency, and enhancing overall reasoning quality of LLMs. Across various mathematical benchmarks, the method delivers consistent gains over strong baselines with standard RL post-training. Specifically, on AIME24, we improve DeepSeek-R1-Distill-Qwen-7B from 54.0 to 61.3 (+7.3) and DeepSeek-R1-Distill-Llama-8B from 43.7 to 53.7 (+10.0). The modular discriminator also enables flexible reward shaping for objectives such as teacher distillation, preference alignment, and mathematical proof-based reasoning.

Applications · Language, Speech and Dialog

Zhiwei He, Tian Liang, Jiahao Xu, Qiuzhi Liu, Xingyu Chen, Yue Wang, Linfeng Song, Dian Yu, Zhenwen Liang, Wenxuan Wang 等

Reinforcement learning (RL) with large language models shows promise in complex reasoning. However, its progress is hindered by the lack of large-scale training data that is sufficiently challenging, contamination-free and verifiable. To solve this problem, we introduce DeepMath-103K, a large-scale mathematical dataset designed with high difficulty (primarily levels 5-9), rigorous decontamination against numerous benchmarks, and verifiable answers for rule-based RL reward. It further includes three distinct R1 solutions adaptable for diverse training paradigms such as supervised fine-tuning. Spanning a wide range of mathematical topics, DeepMath-103K fosters the development of generalizable and advancing reasoning. Notably, models trained on DeepMath-103K achieve leading results on challenging mathematical benchmarks and demonstrate generalization beyond math such as biology, physics and chemistry, underscoring its broad efficacy.

Social Aspects · Fairness, Equity, Justice and Safety

Jiyuan Fu, Kaixun Jiang, Lingyi Hong, Jinglun Li, HaiJing Guo, Dingkang Yang, Zhaoyu Chen, Wenqiang Zhang

Multimodal Large Language Models (MLLMs) have shown great promise but require substantial computational resources during inference. Attackers can exploit this by inducing excessive output, leading to resource exhaustion and service degradation. Prior energy-latency attacks aim to increase generation time by broadly shifting the output token distribution away from the EOS token, but they neglect the influence of token-level Part-of-Speech (POS) characteristics on EOS and sentence-level structural patterns on output counts, limiting their efficacy. To address this, we propose \textbf{LingoLoop}, an attack designed to induce MLLMs to generate excessively verbose and repetitive sequences. First, we find that the POS tag of a token strongly affects the likelihood of generating an EOS token. Based on this insight, we propose a \textbf{POS-Aware Delay Mechanism} to postpone EOS token generation by adjusting attention weights guided by POS information. Second, we identify that constraining output diversity to induce repetitive loops is effective for sustained generation. We introduce a \textbf{Generative Path Pruning Mechanism} that limits the magnitude of hidden states, encouraging the model to produce persistent loops. Extensive experiments on models like Qwen2.5-VL-3B demonstrate LingoLoop's powerful ability to trap them in generative loops; it consistently drives them to their generation limits and, when those limits are relaxed, can induce outputs with up to \textbf{367$\times$} more tokens than clean inputs, triggering a commensurate surge in energy consumption. These findings expose significant MLLMs' vulnerabilities, posing challenges for their reliable deployment.

Applications · Language, Speech and Dialog

Wei Han, Pan Zhou, Shuicheng YAN

The limited context window of contemporary large language models (LLMs) hinders broader application. In this work, we present SharedLLM, a novel approach grounded in the design philosophy of multi-grained context compression and query-aware information retrieval. SharedLLM is composed of two short-context LLMs: a lower moel (compressor) and an upper model (decoder). The lower model compresses context information, while the upper model processes compressed, context information from the lower model and performs context-aware modeling. Information transfer between the compressor and decoder occurs only at the lowest layers to reduce redundant computation. Based on this architecture, we introduce a specialized tree-style data structure to efficiently encode, store and retrieve multi-grained contextual information from text chunks. This entire process, wherein the sender and receiver are derived from the same LLM layer, is referred to as self-injection. In our evaluation on long-context modeling and understanding tasks, SharedLLM achieves superior or comparable results to several strong baselines, striking an effective balance between efficiency and performance. Meanwhile, with the aforementioned design choices, SharedLLM can greatly reduce memory consumption, and demonstrates substantial speed-ups over other advanced baselines. The core code of our implementation along with training and evaluation is available in appendix and supplementary.

Applications · Language, Speech and Dialog

Luyao Zhuang, Shengyuan Chen, Yilin Xiao, Huachi Zhou, Yujing Zhang, Hao Chen, Qinggang Zhang, Xiao Huang

Retrieval-Augmented Generation (RAG) is widely used to mitigate hallucinations of Large Language Models (LLMs) by leveraging external knowledge. While effective for simple queries, traditional RAG systems struggle with large-scale, unstructured corpora where information is fragmented. Recent advances incorporate knowledge graphs to capture relational structures, enabling more comprehensive retrieval for complex, multi-hop reasoning tasks. However, existing graph-based RAG (GraphRAG) methods rely on unstable and costly relation extraction for graph construction, often producing noisy graphs with incorrect or inconsistent relations that degrade retrieval quality. In this paper, we revisit the pipeline of existing GraphRAG systems and propose Linear Graph-based Retrieval-Augmented Generation (LinearRAG), an efficient framework that enables reliable graph construction and precise passage retrieval. Specifically, LinearRAG constructs a relation-free hierarchical graph, termed Tri-Graph, using only lightweight entity extraction and semantic linking, avoiding unstable relation modeling. This new paradigm of graph construction scales linearly with corpus size and incurs no extra token consumption, providing an economical and reliable indexing of the original passages. For retrieval, LinearRAG adopts a two-stage strategy: (i) relevant entity activation via local semantic bridging, followed by (ii) passage retrieval through global importance aggregation. Extensive experiments on four benchmark datasets demonstrate that LinearRAG significantly outperforms baseline models. Our code and datasets are available at https://github.com/DEEP-PolyU/LinearRAG.

Applications · Language, Speech and Dialog

Jihye Choi, Jinsung Yoon, Jiefeng Chen, Somesh Jha, Tomas Pfister

While Large Language Models (LLMs) have shown remarkable advancements in reasoning and tool use, they often fail to generate optimal, grounded solutions under complex constraints. Real-world travel planning exemplifies these challenges, evaluating agents’ abilities to handle constraints that are explicit, implicit, and even evolving based on interactions with dynamic environments and user needs. In this paper, we present ATLAS, a general multi-agent framework designed to effectively handle such complex nature of constraints awareness in real-world travel planning tasks. ATLAS introduces a principled approach to address the fundamental challenges of constraint-aware planning through dedicated mechanisms for dynamic constraint management, iterative plan critique, and adaptive interleaved search. ATLAS demonstrates state-of-the-art performance on the TravelPlanner benchmark, improving the final pass rate from 23.3% to 44.4% over its best alternative. More importantly, our work is the first to demonstrate quantitative effectiveness on real-world travel planning tasks with live information search and multi-turn feedback. In this realistic setting, ATLAS showcases its superior overall planning performance, achieving an 84% final pass rate which significantly outperforms baselines including ReAct (59%) and a monolithic agent (27%).

Applications · Language, Speech and Dialog

Qinglong Yang, Haoming Li, Haotian Zhao, Xiaokai Yan, Jingtao Ding, Fengli Xu, Yong Li

Mobile GUI agents are becoming critical tools to improve user experience on smart devices, with multimodal large language models (MLLMs) emerging as the dominant paradigms in this domain. Current agents, however, rely on explicit human instructions, overlooking the potential to leverage the contextual information (like location, time, user profile) and historical data for proactive task suggestions. Besides, previous works focus on optimizing the success rate during task execution, but pay less attention to the personalized execution trajectory, thereby neglecting potentially vast differences in user preferences. To address these challenges, we introduce the FingerTip 20K benchmark. We collected 20K unique human demonstrations of multi-step Android device interactions across a variety of everyday apps. These demonstrations are not isolated but are continuously acquired from the users' long-term usage in their real lives, and encompass essential user-related contextual information. The benchmark contains two new tracks: proactive task suggestions by analyzing environment observation and users' previous intents, and personalized task execution by catering to users' action preferences. Our experiments reveal that the tracks we propose pose significant challenges for leveraging user-related information in GUI tasks. We also performed a human study to show that there exists a huge gap between existing agents and humans. The model fine-tuned with the data we collected effectively utilized user information and achieved good results, highlighting the potential of our approach in building more user-oriented mobile LLM agents. Our code is open-source at \url{https://github.com/tsinghua-fib-lab/FingerTip-20K} for reproducibility.

Sonia Murthy, Rosie Zhao, Jennifer Hu, Sham Kakade, Markus Wulfmeier, Peng Qian, Tomer Ullman

Value trade-offs are an integral part of human decision-making and language use, however, current tools for interpreting such dynamic and multi-faceted notions of values in language models are limited. In cognitive science, so-called "cognitive models" provide formal accounts of such trade-offs in humans, by modeling the weighting of a speaker's competing utility functions in choosing an action or utterance. Here, we show that a leading cognitive model of polite speech can be used to systematically evaluate alignment-relevant trade-offs in language models via two encompassing settings: degrees of reasoning "effort" and system prompt manipulations in closed-source frontier models, and RL post-training dynamics of open-source models. Our results show that LLMs' behavioral profiles under the cognitive model a) shift predictably when they are prompted to prioritize certain goals, b) are amplified by a small reasoning budget, and c) can be used to diagnose other social behaviors such as sycophancy. Our findings from LLMs' post-training dynamics reveal large shifts in values early on in training and persistent effects of the choice of base model and pretraining data, compared to feedback dataset or alignment method. Our framework offers a flexible tool for probing behavioral profiles across diverse model types and gaining insights for shaping training regimes that better control trade-offs between values during model development.

Applications · Language, Speech and Dialog

Kaicheng Wang, Liyan Huang, Weike Fang, Weihang Wang

Efficient code retrieval is critical for developer productivity, yet existing benchmarks largely focus on Python and rarely stress-test robustness beyond superficial lexical cues. To address the gap, we introduce an automated pipeline for code search datasets and present CLARC, a C/C++ benchmark built from real-world GitHub repositories. CLARC contains 1,245 query-code pairs for evaluation and 5,472 pairs for training. The benchmark incorporates LLM-generated natural language queries validated through rigorous human scoring and hypothesis testing. To analyze contextual requirements effectively, our pipeline starts by ensuring code compilability. It then categorizes code snippets by dependency complexity, distinguishing whether the code relies on custom-defined types or helper functions. The pipeline also enables CLARC to stress-test retrieval robustness by introducing challenging settings, including identifier anonymization and compilation to low-level languages like Assembly and WebAssembly. Under these conditions, our evaluation of six state-of-the-art models reveals sharp drops in retrieval effectiveness. The experimental results highlight the models' persistent reliance on lexical features rather than code semantic understanding. Our dataset is publicly available at https://huggingface.co/datasets/ClarcTeam/CLARC.