论文检索

Yujun Hu, Xiaoyu Zhou, Changbo Wang, Gaoqi He

Diagnosing fine-grained hallucinations in Large Vision-Language Models (LVLMs) can greatly advance their reliable deployment in real-world applications. Nevertheless, current benchmarks predominantly employ flat metrics that treat errors in isolation, leaving a gap in evaluating the complex causal dependencies between visual perception and textual reasoning. Motivated by this, we introduce the Knowledge-Guided In-Context Probing (K-GIP) framework to fill this gap. Specifically, K-GIP constructs a high-fidelity dual-perception ground truth to transform abstract priors into multi-granularity queries. Furthermore, we propose a Verification Scene Graph metric equipped with a Sequential Logic Pruning protocol, which explicitly models existence-attribute dependencies to strictly penalize logical fractures. We conduct comprehensive evaluations of mainstream LVLMs across three datasets using K-GIP. The experimental results highlight that our methodology successfully isolates deep reasoning failures from simple perceptual misses. We hope K-GIP can serve as a valuable and rigorous standard to assess logical robustness in multimodal systems.

Sebastian Antony Joseph, Lily Chen, Barry Wei, Michael Mackert, Iain James Marshall, Paul Pu Liang, Ramez Kouzy, Byron C Wallace, Junyi Jessy Li

Technological progress has led to concrete advancements in tasks that were regarded as challenging, such as automatic fact-checking. Interest in adopting these systems for public health and medicine has grown due to the high-stakes nature of medical decisions and challenges in critically appraising a vast and diverse medical literature. Evidence-based medicine connects to every individual, and yet the nature of it is highly technical, rendering the medical literacy of majority users inadequate to sufficiently navigate the domain. Such problems with medical communication ripen the ground for end-to-end fact-checking agents: check a claim against current medical literature and return with an evidence-backed verdict. And yet, such systems remain largely unused.In this position paper, developed with expert input, we present the first study examining how clinical experts verify real claims from social media by synthesizing medical evidence. In searching for this upper-bound, we reveal fundamental challenges in end-to-end fact-checking when applied to medicine: Difficulties connecting claims in the wild to scientific evidence in the form of clinical trials; ambiguities in underspecified claims mixed with mismatched intentions; and inherently subjective veracity labels. We argue that fact-checking should be approached as an interactive communication problem, rather than an end-to-end process.

Manoj Saravanan, Rohit Kumar Salla, Ramya Manasa Amancherla

Retrieval-Augmented Generation (RAG) systems depend on non-parametric indices to access external knowledge, yet most retrieval infrastructure assumes a stationary query document distribution after index construction. In dynamic settings involving continual knowledge updates or evolving terminology, this assumption often fails, leading to degraded retrieval performance, while full re-indexing remains computationally expensive. We propose AURORA, a neuro-symbolic framework for adapting retrieval indices under distribution shift by treating index maintenance as a few-shot continual learning problem. AURORA decouples discrete index structure from continuous metric representations, enabling efficient adaptation of neural components while preserving index topology. A lightweight Bayesian routing policy further balances stability and plasticity by dynamically selecting among adaptive neural indices and static fallbacks based on uncertainty estimates. Across dense, learned sparse (SPLADE), and generative (DSI) retrieval settings, AURORA recovers up to +26.9% Recall@10 on novel topics compared to static baselines, while adapting significantly faster than full retraining (28 ms vs. 5.1 s).

Yuxuan Zhu, David H. Yang, Mohammad Mohammadi Amiri, Keerthiram Murugesan, Tejaswini Pedapati, Pin-Yu Chen

The expanding long-context capabilities of large language models are constrained by a significant memory bottleneck: the key-value (KV) cache required for autoregressive generation. This bottleneck is substantial; for instance, a Llama-3.1-8B model processing a 32K-token prompt at a batch size of 4 requires approximately 16 GB for its KV cache, exceeding the model’s weights. While KV-cache compression via low-rank projection is promising, existing methods rely on a static, offline-learned subspace that performs poorly under distribution shifts. To overcome these limitations, we introduce OjaKV, a novel framework integrating a hybrid storage policy with online subspace adaptation. OjaKV preserves crucial tokens in full rank as high-fidelity anchors, while applying low-rank compression to intermediate tokens by adapting the projection basis using Oja’s algorithm for online PCA. This adaptation involves a comprehensive update during prefilling and lightweight periodic updates during decoding, ensuring the subspace remains aligned with evolving context. Our framework is fully compatible with FlashAttention. Experiments demonstrate that OjaKV maintains or improves zero-shot accuracy at high compression ratios, achieving the strongest gains on long-context benchmarks requiring complex reasoning. Furthermore, our approach combines with token-selection methods for compounded memory savings, establishing a practical, plug-and-play solution for memory-efficient long-context inference without fine-tuning.

Hayden Helm, Tianyi Chen, Harvey McGuinness, Paige Lee, Brandon Duderstadt, Carey Priebe

In this paper we provide evidence that our virtual model of U.S. congresspersons based on a collection of language models moves towards satisfying the definition of a digital twin. In particular, we introduce and provide high-level descriptions of a daily-updated dataset that contains every Tweet from every U.S. congressperson during their respective terms. We demonstrate that a modern language model equipped with congressperson-specific subsets of this data producing Tweets that are largely indistinguishable from actual Tweets posted by their physical counterparts. We illustrate how generated Tweets can be used to predict roll-call vote behaviors and to quantify the likelihood of congresspersons crossing party lines, thereby assisting stakeholders in allocating resources and potentially impacting real-world legislative dynamics. We conclude with a discussion of the limitations and important extensions of our analysis.

Sachin Kumar

Can small language models achieve strong tool-use performance without complex adaptation mechanisms? This paper investigates this question through Meta-Tool, a controlled empirical study comparing hypernetwork-based LoRA adaptation against carefully designed few-shot prompting. Using a Llama-3.2-3B-Instruct backbone, we evaluate four adaptation mechanisms—few-shot prompting, documentation encoding, hypernetwork-generated LoRA weights, and value-guided beam search—across four diverse benchmarks: Gorilla APIBench, Spider 2.0, WebArena, and InterCode. Our central finding is a well-supported negative result: despite generating non-trivial weight matrices, the 227.8M-parameter hypernetwork provides no measurable improvement over few-shot prompting alone. Comprehensive ablation studies reveal that few-shot examples contribute +21.5% to performance and documentation contributes +5.0%, while the hypernetwork adds 0%. A 3B model with well-designed prompts achieves 79.7% of GPT-5’s average performance at 10 \times lower latency. Error analysis across 722 failure cases spanning all shot counts (0–5) shows that at the 5-shot configuration (106 failures), failure modes are task-dependent: schema-heavy tasks (Spider 2.0, WebArena) show near-zero format errors with remaining failures semantic, while format errors dominate on Gorilla (100%) and InterCode (70%). These findings redirect practitioners toward prompt engineering and example curation rather than complex adaptation architectures.

Neda Jamshidi, Anders Søgaard, Monica Bianchini

Large language models (LLMs) exhibit cultural biases, yet existing benchmarks rely on closed-form, domain-specific questionnaires. We introduce FRAMENET-CULTURES, a benchmark for evaluating cultural alignment in LLMs based on Fillmore-style frame semantics. Using the EveryCulture encyclopedia, we construct a lexicon of 18 cultural frames (e.g., greeting,child-rearing) across 20 countries, treating it as a structured reference for comparison rather than a definitive representation of contemporary societies. For each frame, we prompt five major LLMs—ChatGPT-5, Gemini-2.5-Flash, Mistral-Large, Qwen-3-Max, DeepSeek-V3.2—three times to generate open-ended instantiations, which are manually annotated and binarized. We measure alignment with country- and continent-level profiles using normalized Hamming distance, and validate cultural recognizability through human evaluation of generated dialogues. Under culture-neutral prompting, outputs align most closely with European profiles, followed by Asian and American ones, indicating a consistent cross-model pattern. With culture-specific prompting, models shift toward the target regions, aligning most strongly with Africa for Ethiopia and with Asia for India. FRAMENET-CULTURES is the first open-ended benchmark for cultural alignment relying on frame semantics. Data, prompts, and annotations are publicly available at https://github.com/neda-jamshidi/FrameNet-Cultures.

Yuanchen Wu, Saurabh Verma, Justin Lee, Fangzhou Xiong, Poppy Zhang, Amel Awadelkarim, Xu Chen, Yubai Yuan, Shawndra Hill

Large language models (LLMs) are highly sensitive to prompts, but most automatic prompt optimization (APO) methods assume access to ground-truth references (e.g., labeled validation data) that are costly to obtain. We propose the Prompt Duel Optimizer (PDO), a sample-efficient framework for label-free prompt optimization based on pairwise preference feedback from an LLM judge. PDO casts prompt selection as a dueling-bandit problem and combines (i) Double Thompson Sampling to prioritize informative comparisons under a fixed judge budget, with (ii) top-performer guided mutation to expand the candidate pool while pruning weak prompts. Experiments on BIG-bench Hard (BBH) and MS MARCO show that PDO consistently identifies stronger prompts than label-free baselines, while offering favorable quality–cost trade-offs under constrained comparison budgets.

Iago Alves Brito, Walcy Rios, Julia Soares Dollis, Diogo Fernandes Costa Silva, Arlindo Rodrigues Galvão Filho

Current safety evaluations of large language models (LLMs) create a dangerous illusion of universal protection by aggregating harms under generic categories such as "Identity Hate", obscuring vulnerabilities toward specific populations. In this work, we expose the Selective Safety Trap: a systemic failure mode where models robustly defend specific populations while leaving underrepresented communities highly vulnerable to identical adversarial attacks. To systematically audit this phenomenon, we introduce MiJaBench, a bilingual (English–Portuguese) adversarial benchmark comprising 43,961 controlled jailbreaking prompts across 16 minority groups. By evaluating 14 state-of-the-art LLMs on MiJaBench, we curate 615,454 prompt-response pairs that compose MiJaBench-Align, revealing that safety alignment is not a uniform semantic capability but a demographic hierarchy, with defense rates fluctuating by up to 42% within the same model solely based on the target group. This disparity persists across architectures and languages and is amplified by scaling, indicating that current alignment methods learn group-specific safeguards rather than a generalized notion of harm. Through targeted direct preference optimization (DPO) on a 1B-parameter baseline, we achieve strong zero-shot safety generalizations to entirely unseen demographics and complex attack strategies. We release all datasets and scripts to provide the community with a concrete pathway toward equitable, transferable safety alignment.

Wenbo Zhang, Hengrui Cai, Wenyu Chen

Large language models (LLMs) have demonstrated significant utility in real-world applications, exhibiting impressive capabilities in natural language processing and understanding. Benchmark evaluations are crucial for assessing the capabilities of LLMs as they can provide a comprehensive assessment of their strengths and weaknesses. However, current evaluation methods often overlook the inherent randomness of LLMs by employing deterministic generation strategies or relying on a single random sample, resulting in unaccounted sampling variance and unreliable benchmark score estimates. In this paper, we propose a hierarchical statistical model that provides a more comprehensive representation of the benchmarking process by incorporating both benchmark characteristics and LLM randomness. We show that leveraging multiple generations improves the accuracy of estimating the benchmark score and reduces variance. Multiple generations also allow us to define \mathbb{P} (correct), a prompt-level difficulty score based on correct ratios, providing fine-grained insights into individual prompts. Additionally, we create a data map that visualizes difficulty and semantics of prompts, enabling error detection and quality control in benchmark construction.

Jian Yang, Shuyue Guo, Linzheng Chai, Wei Zhang, Aishan Liu, Chuan Hao, Zhoujun Li, Xin Zhao, Xianglong Liu, Weifeng Lv 等

Large language models (LLMs) are powerful but costly to train, with scaling laws predicting performance from model size, data, and compute. However, different programming languages (PLs) have varying impacts during pre-training that significantly affect base model performance, leading to inaccurate performance prediction. Existing works focus on language-agnostic settings, neglecting the inherently multilingual nature of modern software development. Therefore, it is first necessary to investigate the scaling laws of different PLs, and then consider their mutual influences to arrive at the final multilingual scaling law. In this paper, we present the first systematic exploration of scaling laws for multilingual code pre-training, conducting over 1000+ experiments (Equivalent to 336,000+ H800 hours) across multiple PLs, model sizes (0.2B to 14B parameters), and dataset sizes (1T tokens). We establish scaling laws for code LLMs across multiple programming languages, showing that interpreted languages benefit more from increased scale than compiled ones. Multilingual pre-training provides synergistic benefits, especially between syntactically similar languages, with parallel pairing (concatenating code with translations) significantly enhancing cross-lingual abilities. We propose a proportion-dependent multilingual scaling law that optimally allocates training tokens by prioritizing high-utility languages (e.g., Python), balancing high-synergy pairs (e.g., JavaScript-TypeScript), and reducing allocation to fast-saturating languages (e.g., Rust), achieving superior performance across all languages compared to uniform distribution.

Yahan Zheng, John J. Guerrerio, Soroush Vosoughi, Weicheng Ma

Preprocessing-based methods for stereotype mitigation, such as pre-/post-training on debiased corpora, are widely used in NLP. While these approaches reduce measurable stereotypes for targeted groups, we find they often induce unintended shifts: stereotyping or counter-stereotyping can increase for other demographics, including across unrelated categories. We demonstrate these side effects across two model families (encoder-only and decoder-only), multiple preprocessing strategies (removing stereotypical sentences, removing group mentions, and swapping references), and both pre- and post-training at different data scales on Wikipedia. Standard benchmarks frequently miss these shifts. Using attention-rollout analysis, we observe that such side effects are not accompanied by large changes in attention flow, complicating mechanistic explanations. We discuss implications for evaluation, provide actionable diagnostics, and argue for side-effect-aware, transparent mitigation practices that make claims calibrated to uncertainty.

Yuntao Du, Zitao Li, Bolin Ding, Yaliang Li, Hanshen Xiao, Jingren Zhou, Ninghui Li

Impressive progress has been made in automated problem-solving by the collaboration of large language model (LLM) based agents. However, these automated capabilities also open avenues for malicious applications. In this paper, we study a new threat that LLMs pose to online pseudonymity, called automated profile inference, where an adversary can instruct LLMs to automatically collect and extract sensitive personal attributes from publicly available user activities on pseudonymous platforms. We also introduce an automated profiling framework called AutoProfiler to demonstrate and assess the feasibility of such attacks in real-world scenarios. AutoProfiler consists of four specialized LLM agents that work collaboratively to retrieve and process user online activities and generate a profile with extracted personal information. Experimental results on two real-world datasets and one synthetic dataset show that AutoProfiler is highly effective and efficient, and the inferred attributes are both identifiable and sensitive, posing significant privacy risks. We explore mitigation strategies from different perspectives and advocate for increased public awareness of this emerging privacy threat.

Zhenjiang Mao, Anirudhh Venkat, Artem Bisliouk, Sindhura Kumbakonam Subramanian, Akshat Kothiyal, Saithej Singhu, Ivan Ruchkin

Large Language Models (LLMs) increasingly rely on long-form, multi-step reasoning to solve complex tasks such as mathematical problem solving and scientific question answering. Despite strong performance, existing confidence estimation methods typically reduce an entire reasoning process to a single scalar score, ignoring how confidence evolves throughout the generation. As a result, these methods are often sensitive to superficial factors such as response length or verbosity, and struggle to distinguish correct reasoning from confidently stated errors. We propose to characterize the stepwise confidence signal using Signal Temporal Logic (STL). Using a discriminative STL mining procedure, we discover temporal formulas that distinguish confidence signals of correct and incorrect responses. Our analysis found that the STL patterns generalize across tasks, and numeric parameters exhibit sensitivity to individual questions. Based on these insights, we develop a confidence estimation approach that informs STL blocks with parameter hypernetworks. Experiments on multiple reasoning tasks show our confidence scores are more calibrated than the baselines.

Xiaowen Ma, Yunpu Ma, Chenyang Lin, Sikuan Yan, Jinhe Bi, Zixuan Cao, Yijun Tian, Volker Tresp, Hinrich Schuetze

Leveraging multiple Large Language Models (LLMs) has proven effective for addressing complex, high-dimensional tasks, but current approaches often rely on static, manually engineered multi-agent configurations. To overcome these constraints, we present the Agentic Neural Network (\mathcal{ANN}), a framework that conceptualizes multi-agent collaboration as a layered neural network architecture. In this design, each agent operates as a node, and each layer forms a cooperative team focused on a specific subtask. The proposed framework follows a two-phase optimization strategy: (1) Forward Phase - Drawing inspiration from neural network forward passes, tasks are dynamically decomposed into subtasks, and cooperative agent teams with suitable aggregation methods are constructed layer by layer. (2) Backward Phase - Mirroring backpropagation, we refine both global and local collaboration through iterative feedback, allowing agents to self-evolve their roles, prompts, and coordination. This neuro-symbolic approach enables our framework to create new or specialized agent teams post-training, delivering notable gains in accuracy and adaptability. Across seven benchmark datasets, \mathcal{ANN} surpasses leading multi-agent baselines under the same configurations, showing consistent performance improvements.

Tunazzina Islam

Unsupervised methods are widely used to induce latent semantic structure from large text collections, yet their outputs often contain incoherent, redundant, or poorly grounded clusters that are difficult to validate without labeled data. We propose a **reasoning-based refinement framework** that leverages large language models (LLMs) not as embedding generators, but as semantic judges that validate and restructure the outputs of arbitrary unsupervised clustering algorithms. Our framework introduces three reasoning stages: (i) **coherence verification**, where LLMs assess whether cluster summaries are supported by their member texts; (ii) **redundancy adjudication**, where candidate clusters are merged or rejected based on semantic overlap; and (iii) **label grounding**, where clusters are assigned interpretable labels through a two-stage process that generates and consolidates semantically similar labels in a fully unsupervised manner. This design decouples representation learning from structural validation and mitigates the common failure modes of embedding-only approaches. We evaluate the framework in real-world social media corpora from two platforms with distinct interaction models, demonstrating consistent improvements in cluster coherence and human-aligned labeling quality over classical topic models and recent representation-based baselines. Human evaluation shows strong agreement with LLM-generated labels, despite the absence of gold-standard annotations. We further conduct robustness analysis under matched temporal and volume conditions to assess cross-platform stability. Beyond empirical gains, our results suggest that LLM-based reasoning can serve as a general mechanism for validating and refining unsupervised semantic structure, enabling more reliable and interpretable analysis of large text collections without supervision.

Junyoung Kim, Anton Korikov, Jiazhou Liang, Justin Cui, Yifan Simon Liu, Qianfeng Wen, Mark Zhao, Scott Sanner

While Large Language Models (LLMs) exhibit exceptional zero-shot relevance modeling, their high computational cost necessitates framing passage retrieval as a budget-constrained global optimization problem. Existing approaches passively rely on first-stage dense retrievers, which leads to two limitations: (1) failing to retrieve relevant passages in semantically distinct clusters, and (2) failing to propagate relevance signals to the broader corpus. To address these limitations, we propose Bayesian Active Learning with Gaussian Processes guided by LLM relevance scoring (BAGEL), a novel framework that propagates sparse LLM relevance signals across the embedding space to guide global exploration. BAGEL models the multimodal relevance distribution across the entire embedding space with a query-specific Gaussian Process (GP) based on LLM relevance scores. Subsequently, it iteratively selects passages for scoring by strategically balancing the exploitation of high-confidence regions with the exploration of uncertain areas. Extensive experiments across four benchmark datasets and two LLM backbones demonstrate that BAGEL effectively explores and captures complex relevance distributions and outperforms LLM reranking methods under the same LLM budget on all four datasets.

Xingchen Xiao, Heyan Huang, Runheng Liu, Jincheng Xie

Large language models (LLMs) are widely used in retrieval-augmented generation (RAG) to incorporate external knowledge at inference time. However, when retrieved contexts are noisy, incomplete, or heterogeneous, a single generation process often struggles to reconcile evidence effectively. We propose MASS-RAG, a multi-agent synthesis approach to retrieval-augmented generation that structures evidence processing into multiple role-specialized agents. MASS-RAG applies distinct agents for evidence summarization, evidence extraction, and reasoning over retrieved documents, and combines their outputs through a dedicated synthesis stage to produce the final answer. This design exposes multiple intermediate evidence views, allowing the model to compare and integrate complementary information before answer generation. Experiments on four benchmarks show that MASS-RAG consistently improves performance over strong RAG baselines, particularly in settings where relevant evidence is distributed across retrieved contexts.

Zhiwei Liu, Yupeng Cao, Yuechen Jiang, Mohsinul Kabir, Polydoros Giannouris, Chen Xu, Ziyang Xu, Tianlei Zhu, Md. Tariquzzaman, Triantafillos Papadopoulos 等

Large language models (LLMs) have been widely applied across various domains of finance. Since their training data are largely derived from human-authored corpora, LLMs may inherit a range of human biases. Behavioral biases can lead to instability and uncertainty in decision-making, particularly when processing financial information. However, existing research on LLM bias has mainly focused on direct questioning or simplified, general-purpose settings, with limited consideration of the complex real-world financial environments and high-risk, context-sensitive, multilingual financial misinformation detection tasks (MFMD). In this work, we propose MFMDScen, a comprehensive benchmark for evaluating behavioral biases of LLMs in MFMD across diverse economic scenarios. In collaboration with financial experts, we construct three types of complex financial scenarios: (i) role- and personality-based, (ii) role- and region-based, and (iii) role-based scenarios incorporating ethnicity and religious beliefs. We further develop a multilingual financial misinformation dataset covering English, Chinese, Greek, and Bengali. By integrating these scenarios with misinformation claims, MFMDScen enables a systematic evaluation of 22 mainstream LLMs. Our findings reveal that pronounced behavioral biases persist across both commercial and open-source models. This project is available at https://github.com/lzw108/FMD.

Jucheng Shen, Gaurav Sarkar, Yeonju Ro, Sharath Nittur Sridhar, Zhangyang Wang, Aditya Akella, Souvik Kundu

We present CadLLM, a training-free method to accelerate the inference throughput of diffusion-based LLMs (dLLMs). We first investigate on the dynamic nature of the token unmasking confidence across blocks and steps. Based on this observation, we then present a lightweight adaptive approach that can control the generation block size, step size, and threshold based on the average confidence score of the unmasked tokens. We further reduce the softmaxing overhead of token probability generation by dynamically leveraging a subset of vocabulary size to regulate sampling breadth. CadLLM is a plug-and-play model-agnostic with KV caching based dLLMs. Extensive experiments on four popular tasks demonstrate the efficacy of CadLLM to yield throughput improvement of up to 1.1-2.28x over the state-of-the-art baseline with competitive accuracy.