The rise of Large Audio-Language Models (LAMs) brings both potential and risks, as their audio outputs may contain harmful or unethical content. However, current research lacks a systematic, quantitative evaluation of LAM safety, especially against jailbreak attacks, which are challenging due to the temporal and semantic nature of speech. To bridge this gap, we introduce AJailBench, the first benchmark specifically designed to evaluate jailbreak vulnerabilities in LAMs. We begin by constructing -Base, a dataset of 1,495 adversarial audio prompts spanning 10 policy-violating categories. Using this dataset, we evaluate several state-of-the-art LAMs and reveal that none exhibit consistent robustness across attacks. To further strengthen jailbreak testing and simulate more realistic attack conditions, we propose a method to generate dynamic adversarial variants. Our Audio Perturbation Toolkit (APT) applies targeted distortions across time, frequency, and amplitude domains. To preserve the original jailbreak intent, we enforce a semantic consistency constraint and employ Bayesian optimization to efficiently search for perturbations that are both subtle and highly effective. This results in AJailBench-APT+, an extended dataset of optimized adversarial audio samples. Our findings demonstrate that even small, semantically preserved perturbations can significantly reduce the safety performance of leading LAMs, underscoring the need for more robust and semantically aware defense mechanisms. We release AJailBench to facilitate future research: https://anonymous.4open.science/r/AudioJailbreak-4262/
论文检索
输入标题、作者或关键词，从 9,460 篇学术成果中精准定位
Entity matching (EM) requires fine-grained contextual understanding and domain knowledge. Recent work shows that large language models (LLMs) can serve as strong matchers across domains, but most methods either make independent pairwise decisions or rely on manually designed composite pipelines, thus lacking flexibility in realistic multi-candidate settings. At the same time, they typically ignore inference cost at scale. We formulate LLM-based EM with candidates as a cost-aware sequential decision problem and propose CaRL-EM, a reinforcement learning controller that manages LLM operations. Given the state of an anchor record, its candidate set, and the cost, CaRL-EM adaptively chooses among different operators (Match/Compare/Select/Decide) and model capacities to maximize a quality–cost objective. The policy interacts with abstract operators, allowing the same controller to be reused with different underlying LLM backends at inference time without retraining. Experiments on 7 benchmarks show that CaRL-EM (i) learns to dynamically plan the usage of inexpensive and expensive operators based on task complexity, (ii) achieves robust zero-shot transfer across diverse datasets and domains, and (iii) consistently achieves a better quality–cost trade-off than strong LLM-based baselines and manually designed pipelines, yielding a lower inference cost at comparable or higher quality.
Large language models (LLMs) with long context windows offer the potential to translate entire documents in a single pass, yet they frequently suffer from catastrophic information distortion, undermining the strict faithfulness required for translation. This challenge is compounded by the scarcity of document-level parallel data, which makes both supervised fine-tuning and reliable evaluation prohibitively expensive. We propose LongDu, a self-supervised post-training framework that improves long-document translation reliability via round-trip consistency. Given monolingual documents, LongDu samples multiple candidate translations, back-translates each candidate, and optimizes the model to prefer translations that best reconstruct the source. To make this signal robust for long-form generation, we design a reward that filters trivial failure modes (e.g., copying and local language drift) before applying a reconstruction and fluency score, enabling stable reinforcement learning without human annotations. We additionally introduce Long-CIRT, an automatic evaluation protocol that quantifies information distortion by measuring how much a LLM’s performance degrades after a translation cycle. Across multiple base models, LongDu substantially improves information retention and translation quality, with gains that generalize beyond the training length range and to unseen target languages.
Chain-of-Thought (CoT) reasoning has driven recent gains of large language models (LLMs) on reasoning-intensive tasks by externalizing intermediate steps. However, excessive or redundant reasoning — so-called overthinking — can increase inference costs and lead LLMs toward incorrect conclusions. In this paper, we present \textbf{REFRAIN} (\underline{REF}lective-\underline{R}edundancy for \underline{A}daptive \underline{IN}ference), a training-free framework that adaptively determines when to stop reasoning to mitigate overthinking. REFRAIN integrates a two-stage stop discriminator to identify reflective yet redundant reasoning and a sliding-window Upper Confidence Bound (SW-UCB) multi-armed bandit controller to dynamically adjust stopping thresholds according to problem difficulty without supervision or fine-tuning. Across four representative benchmarks and two model families, REFRAIN reduces token usage by 20-55% while maintaining or improving accuracy compared to standard CoT prompting. Extensive ablation and robustness analyses demonstrate its stability across models, scorers, and prompt variations. In summary, our findings highlight when-to-stop as a new and practical axis of test-time scaling — enabling models to reason not just more, but just enough.
Segment Anything Model 3 (SAM3) advances open-vocabulary segmentation through promptable concept segmentation, enabling users to segment all instances associated with a given concept using short noun-phrase (NP) prompts. While effective for concept-level grounding, real-world interactions often involve far richer natural-language instructions that combine attributes, relations, actions, states, or implicit reasoning. Currently, SAM3 relies on external multi-modal agents to convert complex instructions into NPs and conducts iterative mask filtering, leading to coarse representations and limited instance specificity. In this work, we present SAM3-I, an instruction-following extension of the SAM family that unifies concept-level grounding and instruction-level reasoning within a single segmentation framework. Built upon SAM3, SAM3-I introduces an instruction-aware cascaded adaptation mechanism with dedicated alignment losses that progressively aligns expressive instruction semantics with SAM3’s vision-language representations, enabling direct interpretation of natural-language instructions while preserving its strong concept recall ability. To enable instruction-following learning, we introduce HMPL-Instruct, a large-scale instruction-centric dataset that systematically covers hierarchical instruction semantics and diverse target granularities. Experiments demonstrate that SAM3-I achieves appealing performance across referring and reasoning-based segmentation, showing that SAM3 can be effectively extended to follow complex natural-language instructions without sacrificing its original concept-driven strengths. Code and dataset are available at https://github.com/debby-0527/SAM3-I.
Inscriptions are invaluable cultural heritage, yet centuries of degradation (e.g., fractures, erosion, oxidation) have rendered many partially illegible. Existing Historical Inscription Restoration (HIR) methods rely on task-separated pipelines with irreversible error accumulation and patch-based generation that sacrifices page-level consistency. Therefore, we present UniHIR, the first unified MLLM for end-to-end historical inscription restoration. It integrates two novel designs, Draft-Guided Localization and Hierarchical Self-Refinement, to enable accurate damage localization and illegible-content prediction via iterative reasoning and self-correction. This unified approach enables true page-level restoration with consistent typography and style. To support training under high-resolution inputs and long sequences, we design UHIRFactory and construct HIRBench, enabling step-wise, memory-efficient instruction tuning with step-aware annotations for intermediate drafts and refinements. Experiments demonstrate that UniHIR achieves superior performance in both text restoration accuracy and appearance restoration quality, validating that HIR can be effectively tackled by a standalone model in a unified manner. The model and code are available at https://github.com/ZZXF11/UniHIR.
We introduce Parallel Coordinated Reasoning (PaCoRe), a training-and-inference framework designed to overcome a central limitation of contemporary language models: their inability to scale test-time compute (TTC) far beyond sequential reasoning under a fixed context window. PaCoRe departs from the traditional sequential paradigm by driving TTC through massive parallel exploration coordinated via a message-passing architecture in multiple rounds. Each round launches many parallel reasoning trajectories, compacts their findings into context-bounded messages, and synthesizes these messages to guide the next round and ultimately produce the final answer. Trained end-to-end with large-scale, outcome-based reinforcement learning, the model masters the synthesis abilities required by PaCoRe and scales to multi-million-token effective TTC without exceeding context limits. The approach yields strong improvements across diverse domains and notably pushes reasoning beyond frontier systems in mathematics: an 8B model reaches 94.5% on HMMT 2025, surpassing GPT-5’s 93.2% by scaling effective TTC to roughly two million tokens. We open-source model checkpoints, training data, and the full inference pipeline to accelerate follow-up work.
Large Language Model (LLM) agents struggle with ultra-long-horizon tasks requiring hundreds or thousands of interaction steps. Traditional context management approaches face a fundamental dilemma: preserving complete histories rapidly exhausts context windows and forces crude truncation, while aggressive summarization discards critical information prematurely. We propose Predictive Adaptive Context Extraction (PACE), a novel framework that reconceptualizes context management as a Next Step Prediction problem. Inspired by neural attention, PACE dynamically constructs context by adjusting historical memory granularity based on its predicted relevance for the next action. Comprehensive evaluation across diverse benchmarks and models demonstrates that PACE consistently improves task success rates, with larger gains on complex tasks and robust cross-lingual performance. Crucially, PACE enables agents to sustain effective reasoning for 4,897 interaction steps in ultra-long-horizon scenarios, achieving a 66.2 improvement over the full-context ReAct baseline and 5.1 over advanced folding baselines. This fundamentally advances the capability of LLM-based agents in previously intractable long-horizon scenarios. Our code and data are available at https://anonymous.4open.science/r/PACE-B000/.
LADR: Locality-Aware Dynamic Rescue for Efficient Text-to-Image Generation with Diffusion Large Language Models
PDF ↗Discrete Diffusion Language Models have emerged as a compelling paradigm for unified multimodal generation, yet their deployment is hindered by high inference latency arising from iterative decoding. Existing acceleration strategies often require expensive re-training or fail to leverage the 2D spatial redundancy inherent in visual data. To address this, we propose Locality-Aware Dynamic Rescue (LADR), a training-free method that expedites inference by exploiting the spatial Markov property of images. LADR prioritizes the recovery of tokens at the “generation frontier”, regions spatially adjacent to observed pixels, thereby maximizing information gain. Specifically, our method integrates morphological neighbor identification to locate candidate tokens, employs a risk-bounded filtering mechanism to prevent error propagation, and utilizes manifold-consistent inverse scheduling to align the diffusion trajectory with the accelerated mask density. Extensive experiments on four text-to-image generation benchmarks demonstrate that our LADR achieves an approximate 4 \times speedup over standard baselines. Remarkably, it maintains or even enhances generative fidelity, particularly in spatial reasoning tasks, offering a state-of-the-art trade-off between efficiency and quality.
The increasing context window greatly extends the capabilities of large language models, but on the other hand, it incurs an unaffordable memory overhead and computational latency due to the increasing Key-Value (KV) cache size. Recent KV cache compression methods manage to reduce the cache size by dropping irrelevant KVs. However, these methods often fail to identify crucial KVs for generation while excluding others accurately, resulting in severe information loss. To address this gap, we propose **IntentKV**, an intention-aware KV cache eviction method that identifies and retains crucial KVs according to the attention distribution of intention, which semantically reflects the user’s goal and determines which part of the context is relevant. The consistency between the semantics and attention distribution is further substantiated through meticulously designed experiments. On this basis, IntentKV first distinguishes intention tokens from the vanilla context tokens based on their attention distribution distances. Then, the block-wise cumulative attention is calculated via aggregating the intention token attention. Finally, blocks that acquire high cumulative attention are picked and stored in KV cache. We evaluate our method across diverse long-context tasks and models. Results demonstrate that IntentKV can effectively maintain the model performance while reducing the KV cache size from 128K to 2K, leading to a 6.3x increase in decoding speed and 7.8x enhancement in memory efficiency compared to the default setting.
As Large Language Models (LLMs) increasingly operate as Deep Research (DR) Agents capable of autonomous investigation and information synthesis, reliable evaluation of their task performance has become a critical bottleneck. Current benchmarks predominantly rely on static datasets, which suffer from several limitations: limited task generality, temporal misalignment, and data contamination. To address these, we introduce DR-Arena, a fully automated evaluation framework that pushes DR agents to their capability limits through dynamic investigation. DR-Arena constructs real-time Information Trees from fresh web trends to ensure the evaluation rubric is synchronized with the live world state, and employs an automated Examiner to generate structured tasks testing two orthogonal capabilities: Deep reasoning and Wide coverage. DR-Arena further adopts Adaptive Evolvement Loop, a state-machine controller that dynamically escalates task complexity based on real-time performance, demanding deeper deduction or wider aggregation until a decisive capability boundary emerges. Experiments with six advanced DR agents demonstrate that DR-Arena achieves a Spearman correlation of 0.94 with the LMSYS Search Arena leaderboard. This represents state-of-the-art alignment with human preferences without any manual efforts, validating DR-Arena as a reliable alternative for costly human adjudication.
While natural language is the de facto communication medium for LLM-based agents, it presents a fundamental constraint. The process of downsampling rich, internal latent states into discrete tokens inherently limits the depth and nuance of information that can be transmitted, thereby hindering collaborative problem-solving. Inspired by telepathy, which bypasses symbolic language in communication, we propose Interlat (Inter-agent Latent Space Communication), a paradigm that leverages the continuous last hidden states of an LLM as a representation of its thought for direct communication (termed "latent communication"). An additional learned compression process further compresses latent communication via latent space reasoning. Experiments demonstrate that Interlat outperforms both fine-tuned chain-of-thought (CoT) prompting and single-agent baselines, even across heterogeneous models, promoting more exploratory behavior and enabling genuine utilization of latent information. Further compression not only substantially accelerates inference by up to 24× but also maintains competitive performance through an efficient information-preserving mechanism. We position this work as a feasibility study of entirely latent space inter-agent communication, and our results highlight its potential, offering valuable insights for future research.
LiveCultureBench: a Multi-Agent, Multi-Cultural Benchmark for Large Language Models in Dynamic Social Simulations
PDF ↗Large language models (LLMs) are increasingly deployed as autonomous agents, yet evaluations focus primarily on task success rather than cultural appropriateness or evaluator reliability. We introduce LiveCultureBench, a multi-cultural, dynamic benchmark that embeds LLMs as agents in a simulated town and evaluates them on both task completion and adherence to socio-cultural norms. The simulation models a small city as a location graph with synthetic residents having diverse demographic and cultural profiles. Each episode assigns one resident a daily goal while others provide social context. An LLM-based verifier generates structured judgments on norm violations and task progress, which we aggregate into metrics capturing task-norm trade-offs and verifier uncertainty. Using LiveCultureBench across models and cultural profiles, we study (i) cross-cultural robustness of LLM agents, (ii) how they balance effectiveness against norm sensitivity, and (iii) when LLM-as-a-judge evaluation is reliable for automated benchmarking versus when human oversight is needed.
Large Language Models (LLMs) behave non-deterministically, and prompting has become a common method for steering their outputs.A popular strategy is to assign a persona to the model to produce more varied, context-sensitive responses, similar to how responses vary across human individuals.Against the expectation that persona prompting yields a wide range of opinions, our experiments show that LLMs keep consistent value orientations.We observe a persistent inertia in their responses, where certain moral and value dimensions (especially harm avoidance and fairness) stay skewed in one direction across persona settings.To study this, we use role-play at scale, which pairs randomized persona prompts with a macro-level analysis of model outputs.Our results point to strong internal biases and value preferences in LLMs, which we call value orientation and inertia. These models warrant scrutiny and adjustment before use in applications where balanced outputs matter.
Attention Weights as an Indicator: Analyzing and Improving Document Utilization in Retrieval-Augmented Generation
PDF ↗The generation of Retrieval-Augmented Generation (RAG) models is affected by factors such as the quality and order of input documents, indicating that their ability to utilize documents remains underdeveloped. This ability encompasses not only identifying useful documents from inputs but also minimizing positional bias and filtering irrelevant documents. To achieve this, key challenges include the model’s internal estimation of document importance and positional bias. In this paper, we conduct a comprehensive study on the properties of attention weights, examining the impact of factors like aggregation methods, document quality, document position, token type, and so on. Based on our findings, we propose strategies to enhance document utilization from three perspectives: document ranking, placement, and filtering. Comprehensive experiments show that our method outperforms baselines and improves document utilization effectiveness in a training-free manner.
Harmonizing the Past, Present, and Future: A Null-Space Constrained Region-Specific Method for Continual Learning in LLMs
PDF ↗Enabling Large Language Models (LLMs) to evolve sustainably requires simultaneously preserving previously acquired knowledge (Past), effectively acquiring new task-specific skills (Present), and reserving sufficient parameter capacity for subsequent adaptation (Future). However, existing continual learning (CL) paradigms often prioritize immediate performance through dense updates, leading to catastrophic forgetting and rapid exhaustion of model capacity. To harmonize these conflicting demands, we draw inspiration from the brain’s functional partitioning and propose the Null-Space Constrained Parameter Region Specificity Method (PaRSP). PaRSP establishes a dynamic "Task-Region Mapping" that distinguishes between specialized neurons and generalist neurons. By precisely localizing a sparse "functional core" for each task, PaRSP restricts updates to specific regions via null-space orthogonality, preserving the vast majority of the network as an immutable "long-term memory bank." This induced sparsity not only enhances plasticity via targeted adaptation and minimizes interference to ensure stability, but also strategically reserves substantial capacity, securing sustainability for future evolution. Extensive experiments validate PaRSP’s state-of-the-art performance, particularly on Standard CL and Long Sequence benchmarks, effectively harmonizing the stability-plasticity-sustainability trade-off. Code is available at https://github.com/JinhuiBot/PaRSP
Standardized patients (SPs) are indispensable for clinical skills training but remain expensive and difficult to scale. Although large language model (LLM)-based virtual standardized patients (VSPs) have been proposed as an alternative, their behavior remains unstable and lacks rigorous comparison with human standardized patients. We propose EasyMED, a multi-agent VSP framework that separates case-grounded information disclosure from response generation to support stable, inquiry-conditioned patient behavior. We also introduce SPBench, a human-grounded benchmark with eight expert-defined criteria for interaction-level evaluation. Experiments show that EasyMED more closely matches human SP behavior than existing VSPs, particularly in case consistency and controlled disclosure. A four-week controlled study further demonstrates learning outcomes comparable to human SP training, with stronger early gains for novice learners and improved flexibility, psychological safety, and cost efficiency.
ReCoQA: A Benchmark for Tool-Augmented and Multi-Step Reasoning in Real Estate Question and Answering
PDF ↗Developing agents capable of navigating fragmented, multi-source information remains challenging, primarily due to the scarcity of benchmarks reflecting hybrid workflows combining database querying with external APIs. To bridge this gap, we introduce ReCoQA, a large-scale benchmark of 29,270 real-estate instances featuring machine-verifiable supervision for intermediate steps, including structured intent labels, SQL queries, and API calls. Complementarily, we propose HIRE-Agent, a hierarchical framework instantiating an understand–plan–execute architecture as a strong baseline. By orchestrating a Front-end parser, a planning Supervisor, and execution Specialists, HIRE-Agent effectively integrates heterogeneous evidence. Extensive experiments demonstrate that HIRE-Agent constitutes a strong baseline and substantiates the necessity of hierarchical collaboration for complex, real-world reasoning tasks.
Estimating the cognitive complexity of reading comprehension (RC) items is crucial for assessing item difficulty before it is administered to learners. Unlike syntactic and semantic features, such as passage length or semantic similarity between options, cognitive features that arise during answer reasoning are not readily extractable using existing NLP tools and have traditionally relied on human annotation. In this study, we examine whether large language models (LLMs) can estimate the cognitive complexity of RC items by focusing on two dimensions—Evidence Scope and Transformation Level—that indicate the degree of cognitive burden involved in reasoning about the answer. Our experimental results demonstrate that LLMs can approximate the cognitive complexity of items, indicating their potential as tools for prior difficulty analysis. Further analysis reveals a gap between LLMs’ reasoning ability and their metacognitive awareness: even when they produce correct answers, they sometimes fail to correctly identify the features underlying their own reasoning process.
CoCoNUTS: Concentrating on Content while Neglecting Uninformative Textual Styles for AI-Generated Peer Review Detection
PDF ↗The growing use of large language models (LLMs) in peer review threatens scholarly integrity. Recent conference policies allow AI tools for language polishing but prohibit their use for generating substantive content. However, existing detectors mainly rely on stylistic cues, making it difficult to distinguish between surface-level language refinement and genuine content generation. To address this, we advocate a content-based detection paradigm and introduce CoCoNUTS, a comprehensive benchmark containing 315,535 reviews covering leading AI conferences and six human-AI collaboration modes. Our evaluation shows that current detectors struggle to handle these nuanced settings. Consequently, we propose CoCoDet, an AI review detector designed to identify substantive AI-generation. Experiments demonstrate that CoCoDet achieves a macro F1-score of 98.24%. Crucially, on permissible machine-polished reviews, it maintains a low false positive rate of 3.89%, substantially outperforming the strongest baseline (7.84%). Examination on real-world reviews using CoCoDet reveals an escalating trend of substantive AI generation. Our work exposes the inadequacy of current detectors, underscoring the importance of domain-specific solutions.