Large language models (LLMs) can carry out human-like dialogue, but unlike humans, they are stateless due to the superposition property. However, during multi-turn, multi-agent interactions, LLMs begin to exhibit consistent, character-like behaviors—hinting at a form of emergent lifelong learning. Despite this, existing benchmarks often fail to capture these dynamics, primarily focusing on static, open-ended evaluations. To address this gap, we introduce LifeState-BENCH, a benchmark designed to assess lifelong learning in LLMs. It features two episodic datasets—Hamlet and a synthetic script collection—rich in narrative structure and character interactions. Our fact-checking evaluation probes models’ self-awareness, episodic memory retrieval, and relationship tracking, across both parametric and non-parametric approaches. Experiments on models like Llama3.1-8B, GPT-4-turbo, and DeepSeek R1, we demonstrate that non-parametric methods significantly outperform parametric ones in managing stateful learning. However, all models exhibit challenges with catastrophic forgetting as interactions extend, highlighting the need for further advancements in lifelong learning.
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
Existing document OCR largely targets plain text or Markdown, discarding the structural and executable properties that make LaTeX essential for scientific publishing. We study page-level reconstruction of scientific PDFs into compilable LaTeX and introduce TexOCR-Bench, a benchmark, and TexOCR-Train, a large-scale training corpus, for this task. TexOCR-Bench features a multi-dimensional evaluation suite that jointly assesses transcription fidelity, structural faithfulness, and end-to-end compilability. Leveraging TexOCR-Train, we train a 2B-parameter model, TexOCR, using supervised fine-tuning (SFT) and reinforcement learning (RL) with verifiable rewards derived from LaTeX unit tests that directly enforce compilability and referential integrity. Experiments across 21 frontier models on TexOCR-Bench show that existing systems frequently violate key document invariants, including consistent section structure, correct float placement, and valid label–reference links, which undermines compilation reliability and downstream usability. Our analysis further reveals that RL with verifiable rewards yields consistent improvements over SFT alone, particularly on structural and compilation metrics.
Automated interlinear gloss prediction with neural networks is a promising approach to accelerate language documentation efforts. However, while state-of-the-art models like GlossLM (Ginn et al., 2024) achieve high scores on glossing benchmarks, user studies with linguists have found critical barriers to the usefulness of such models in real-world scenarios (Rice et al., 2025). In particular, existing models typically generate morpheme-level glosses but assign them to whole words without predicting the actual morpheme boundaries, making the predictions less interpretable and thus untrustworthy to human annotators.We conduct the first study on neural models that jointly predict interlinear glosses and the corresponding morphological segmentation from raw text. We run experiments to determine the optimal way to train models that balance segmentation and glossing accuracy, as well as the alignment between the two tasks. We extend the training corpus of GlossLM and pretrain PolyGloss, a family of seq2seq multilingual models for joint segmentation and glossing that outperforms GlossLM on glossing and beats various open-source LLMs on segmentation, glossing, and alignment. In addition, we demonstrate that PolyGloss can be quickly adapted to a new dataset via low-rank adaptation.
Scaling multimodal large language models (MLLMs) to long videos is constrained by limited context windows. While retrieval-augmented generation (RAG) is a promising remedy by organizing query-relevant visual evidence into a compact context, most existing methods (i) flatten videos into independent segments, breaking their inherent spatio-temporal structure, and (ii) depend on explicit semantic matching, which can miss cues that are implicitly relevant to the query’s intent. To overcome these limitations, we propose VideoStir, a structured and intent-aware long-video RAG framework. It firstly structures a video as a spatio-temporal graph at clip level, and then performs multi-hop retrieval to aggregate evidence across distant yet contextually related events. Furthermore, it introduces an MLLM-backed intent-relevance scorer that retrieves frames based on their alignment with the query’s reasoning intent. To support this capability, we curate IR-600K, a large-scale dataset tailored for learning frame–query intent alignment. Experiments show that VideoStir is competitive with state-of-the-art baselines without relying on auxiliary information, highlighting the promise of shifting long-video RAG from flattened semantic matching to structured, intent-aware reasoning. Codes and checkpoints are available at https://github.com/RomGai/VideoStir.
Large Language Models (LLMs) are increasingly adopted as conversational assistants in genomics, where they are mainly used to reason over biological knowledge, annotations, and analysis outputs through natural language interfaces. However, existing benchmarks either focus on specialized DNA models trained for sequence prediction or evaluate biological knowledge using text-only questions, leaving the behavior of general-purpose LLMs when directly exposed to raw genome sequences underexplored. We introduce GenomeQA, a benchmark designed to provide a controlled evaluation setting for general-purpose LLMs on sequence-based genome inference tasks. GenomeQA comprises 5,200 samples drawn from multiple biological databases, with sequence lengths ranging from 6 to 1,000 base pairs (bp), spanning six task families: Enhancer and Promoter Identification, Splice Site Identification, Taxonomic Classification, Histone Mark Prediction, Transcription Factor Binding Site Prediction, and TF Motif Prediction. Across six frontier LLMs, we find that models often outperform random baselines, particularly on tasks driven by local sequence cues such as GC content and short motifs, while performance degrades on tasks that require more indirect or multi-step inference over sequence patterns. GenomeQA establishes a diagnostic benchmark for studying and improving the use of general-purpose LLMs on raw genomic sequences.
Audio MultiChallenge: A Multi-Turn Evaluation of Spoken Dialogue Systems on Natural Human Interaction
PDF ↗End-to-end (E2E) spoken dialogue systems are replacing cascaded pipelines for voice-based human-AI interaction. Existing benchmarks primarily evaluate these systems on synthetic speech and single-turn tasks, leaving multi-turn conversational ability underexplored. We introduce Audio MultiChallenge an open-source benchmark to evaluate these systems under natural multi-turn interaction patterns. Building on the text-based MultiChallenge framework, which evaluates Inference Memory, Instruction Retention, and Self Coherence, we introduce a new axis Voice Editing that tests robustness to mid-utterance speech repairs and backtracking. We augment each axis to the audio modality, such as introducing Audio-Cue challenges for Inference Memory that require recalling ambient sounds and paralinguistic signals beyond semantic content. We curate 452 conversations from 47 speakers with 1,712 instance-specific rubrics through a hybrid pipeline that exposes model failures at scale while preserving natural disfluencies found in unscripted human speech. Our evaluation reveals that even frontier models struggle on our benchmark, with our highest-performing model achieving a 54.65% pass rate. Error analysis shows that models are not sufficiently robust to human speech when tracking instructions, edits, and audio cues, highlighting the need for improved audio-native multi-turn interaction capabilities.
Structured pruning is a practical approach to deploying large language models (LLMs) efficiently, as it yields compact, hardware-friendly architectures. However, the dominant local paradigm is task-agnostic: by optimizing layer-wise reconstruction rather than task objectives, it tends to preserve perplexity or generic zero-shot behavior but fails to capitalize on modest task-specific calibration signals, often yielding limited downstream gains. We revisit global structured pruning and present GISP, *Global Iterative Structured Pruning*, a post-training method that removes attention heads and MLP channels using first-order, loss-based important scores aggregated at the structure level with block-wise normalization. Built on this global importance metric, GISP adopts an iterative schedule, rather than one-shot pruning, stabilizes accuracy at higher sparsity, and mitigates perplexity collapse without requiring intermediate fine-tuning. Importantly, the iterative pruning forms nested subnetworks that support a ”prune-once, deploy-many” workflow. Furthermore, GISP defines structural importance directly with respect to a target loss, making it easy to adapt pruning to task-specific objectives. In this work, we use perplexity for language modeling and a margin-based objective for decision-style tasks. Extensive experiments show that across Llama2-7B/13B, Llama3-8B, and Mistral-0.3-7B, GISP consistently lowers WikiText-2 perplexity and improves downstream accuracy, with especially strong gains at 40–50% sparsity; on DeepSeek-R1-Distill-Llama-3-8B and Qwen3-8B with GSM8K, task-aligned calibration substantially boosts exact-match accuracy.
Large language models (LLMs) exhibit exceptional performance but pose inherent risks of generating toxic content, restricting their safe deployment. While traditional methods (e.g., alignment) adjust output preferences, they fail to eliminate underlying toxic regions in parameters, leaving models vulnerable to adversarial attacks. Prior mechanistic studies characterize toxic regions as "toxic vectors" or "layer-wise subspaces", yet our analysis identifies critical limitations: i) Removed toxic vectors can be reconstructed via linear combinations of non-toxic vectors, demanding targeting of entire toxic subspace; ii) Contrastive objective over limited samples inject noise into layer-wise subspaces, hindering stable extraction. These highlight the challenge of identifying robust toxic subspace and removing them. Therefore, we propose GLOSS (GLobal tOxic Subspace Suppression), a lightweight method that mitigates toxicity by identifying and eliminating this global subspace from FFN parameters. Experiments on LLMs (e.g., Qwen3) show GLOSS achieves SOTA detoxification while preserving general capabilities without requiring large-scale retraining.
Large Language Models enhanced with Retrieval Augmented Generation show strong potential in knowledge intensive tasks. However, they often encounter knowledge conflicts, where retrieved information contradicts the model’s internal knowledge or exhibits internal inconsistencies. Existing methods treat this as a simplistic binary choice, forcing models to blindly trust external contexts or rigidly rely on memory, resulting in unreliable predictions that swing between sycophancy and stubbornness. We argue that a more principled approach is to embrace contradictions as opportunities for deeper reasoning. To this end, we introduce Debate-of-Thoughts (DoT), a framework that transforms conflict resolution into an active deliberation process. DoT guides a single model through three phases: 1) hypothesis generation, which forms competing perspectives; 2) internal debate, where the model acts as both a proponent and a critic to stress test each view; and 3) adjudication, where a judge module evaluates arguments based on evidence and logical consistency. We implement DoT via two complementary strategies: inference time prompt chaining and supervised fine tuning. Experiments across multiple conflict benchmarks show that DoT consistently outperforms state-of-the-art methods, while generating transparent debate transcripts that explain its decisions. By improving both accuracy and interpretability under knowledge conflicts, DoT establishes a more reliable paradigm for retrieval augmented generation systems.
Positional encodings are fundamental to Transformers, yet explicit methods like RoPE can degrade under length extrapolation and may incur extra arithmetic and memory-access overhead. In this paper, we propose Scoped Position Encoding (ScoPE), a novel framework that reimagines structured sparsity as an intrinsic position encoding mechanism. Instead of relying on explicit arithmetic signals, ScoPE assigns exponentially scaled look-back scopes to attention heads. We theoretically demonstrate that this simple topological constraint transforms multi-head attention into a hierarchical processor, yielding an order awareness horizon that grows exponentially with depth up to the sequence length. Consequently, ScoPE is parameter-free and avoids relying on fragile positional arithmetic. Empirically, it significantly enhances efficiency by masking the majority of attention computations, offering a theoretical 8x reduction in attention FLOPs at long contexts. Extensive evaluations on LLaMA-3-8B architectures reveal that ScoPE achieves superior native length extrapolation and robust retrieval fidelity compared to RoPE, all while substantially reducing training and inference latency. The code is available at https://github.com/oncemoe/ScoPE.
SLICEFORMER: Static Program Slicing Using Language Models With Dataflow-Aware Pretraining and Constrained Decoding
PDF ↗Static program slicing is a fundamental software engineering technique for isolating code relevant to specific variables. While recent learning-based approaches using language models (LMs) show promise in automating slice prediction, they suffer from inaccurate dependency modeling and unconstrained generation, where LMs fail to capture precise data flow relations and produce slices containing hallucinated tokens and statements. To address these challenges, we propose SliceFormer, a novel approach that reformulates static program slicing as a sequence-to-sequence task using small language models such as CodeT5+. introduces two key innovations that directly target the identified limitations. First, to improve dependency modeling, we design dataflow-aware pretraining objectives that leverage data flow graphs DFG to teach models data dependencies through dataflow-preserving statement permutation and dataflow-aware span corruption. Second, to eliminate hallucination, we develop a constrained decoding mechanism that enforces both lexical and syntactic constraints. We evaluate SliceFormer on Java and Python program slicing benchmarks, demonstrating consistent improvements over state-of-the-art baselines with up to 22% gain in ExactMatch.
Automated schedule generation for multitask from natural language descriptions has huge potential in modern industry. While classic methods bypass language complexities by using pre-formatted matrices, and recent LLM+solver approaches introduce new fragilities by relying on solver-specific code generation. This raises critical questions: Can large language models (LLMs) solve this NL \Rightarrow Schedule task end-to-end well(RQ1)? If the answer is "no", where do they fall short(RQ2)? And how can their capabilities be enhanced (RQ3)? To answer these questions, we introduce NL \Rightarrow Schedule, the first benchmark for this task, equipped with a dataset of 240 description-schedule pairs constructed from real-world materials and a rigorous evaluation suite. Our evaluation of nine state-of-the-art LLMs reveals the limitations of different LLMs in procedure grounding and the strengths of advanced LLMs in global planning via local analysis. To address these shortcomings, we propose Mans, a novel multi-agent framework. Extensive experiments show that Mans achieves more robust performance comparable to six state-of-the-art LLM+solver methods. We hope NL \Rightarrow Schedule and Mans will serve as a solid foundation for automatic scheduling.
Large language model (LLM)-based agents are increasingly deployed in e-commerce shopping. To perform thorough, user-tailored product searches, agents should interpret personal preferences, engage in multi-turn dialogues, and ultimately retrieve and discriminate among highly similar products. However, existing research has yet to provide a unified simulation environment that consistently captures all of these aspects, and always focuses solely on evaluation benchmarks without training support. In this paper, we introduce ShopSimulator, a large-scale and challenging Chinese shopping environment. Leveraging ShopSimulator, we evaluate LLMs across diverse scenarios, finding that even the best-performing models achieve less than 40% full-success rate. Error analysis reveals that agents struggle with deep search and product selection in long trajectories, fail to balance the use of personalization cues, and to effectively engage with users. Further training exploration provides practical guidance for overcoming these weaknesses, with the combination of supervised fine-tuning (SFT) and reinforcement learning (RL) yielding significant performance improvements.
Conversational recommender systems (CRSs) operate under incremental preference revelation, requiring recommendation decisions under uncertainty. While recent LLM-based approaches achieve strong performance on proxy metrics such as Recall@K and BLEU, they often fail to deliver high-quality, user-aligned recommendations in practice, as they optimize intermediate objectives like retrieval accuracy or fluent generation rather than recommendation quality itself. We propose HARPO (Hierarchical Agentic Reasoning with Preference Optimization), an agentic framework that reframes conversational recommendation as a structured decision-making process optimized for multi-dimensional recommendation quality. HARPO integrates (i) hierarchical preference learning that decomposes recommendation quality into interpretable dimensions (relevance, diversity, satisfaction, and engagement) with context-dependent weighting; (ii) deliberative tree-search reasoning guided by a learned value network evaluating candidate paths on predicted quality; and (iii) domain-agnostic reasoning abstractions through Virtual Tool Operations and multi-agent refinement. We evaluate HARPO on ReDial, INSPIRED, and MUSE, demonstrating consistent improvements over strong baselines on recommendation-centric metrics while maintaining competitive response quality.
User Perceptions vs. Proxy LLM Judges: Privacy and Helpfulness in LLM Responses to Privacy-Sensitive Scenarios
PDF ↗Large language models (LLMs) are rapidly being adopted for tasks like draftingemails, summarizing meetings, and answering health questions. In thesesettings, users may need to share private information (e.g., contactdetails, health records). To evaluate LLMs’ ability to identify and redactsuch information, prior work introduced real-life, scenario-based benchmarks(e.g., ConfAIde, PrivacyLens) and found that LLMs can leak privateinformation in complex scenarios. However, these evaluations relied on proxy LLMs to judge the helpfulnessand privacy-preservation quality of LLM responses, rather than directlymeasuring users’ perceptions. To understand how users perceive the helpfulness and privacy-preservationquality of LLM responses to privacy-sensitive scenarios, we conducted auser study (n=94) using 90 PrivacyLens scenarios. We found that users hadlow agreement with each other when evaluating identical LLM responses. Incontrast, five proxy LLMs reached high agreement, yet each proxy LLM hadlow correlation with users’ evaluations. These results indicate that proxy LLMs cannot accurately estimate users’ wide range of perceptions of utility and privacy inprivacy-sensitive scenarios. We discuss the need for more user-centeredstudies to measure LLMs’ ability to help users while preserving privacy,and for improving alignment between LLMs and users in estimating perceivedprivacy and utility.
Video temporal understanding is crucial for multimodal large language models (MLLMs) to reason over events in videos. Despite recent advances in general video understanding, current MLLMs still struggle with fine-grained temporal reasoning. While reinforcement learning (RL) has been explored to address this issue recently, existing RL approaches remain limited in performance on time-sensitive tasks. In this work, we propose **MUSEG**, a novel RL-based method that enhances temporal understanding by introducing timestamp-aware multi-segment grounding. MUSEG enables MLLMs to align queries with multiple relevant video segments, promoting more comprehensive temporal reasoning. To facilitate effective learning, we design a customized RL training recipe with phased rewards that progressively guides the model toward temporally grounded reasoning. Extensive experiments on temporal grounding and time-sensitive video question answering (QA) tasks demonstrate that MUSEG significantly outperforms existing methods and generalizes well across diverse temporal understanding scenarios.
Enzyme–reaction retrieval is a fundamental problem in computational biology, underpinning enzyme characterization, reaction mechanism elucidation, and the rational design of metabolic pathways and biocatalysts. As a bidirectional task, it entails both enzyme-to-reaction and reaction-to-enzyme mapping. However, existing approaches suffer from poor generalization across tasks and distributions, with performance highly sensitive to dataset splits and substantial asymmetry between retrieval directions. To address these challenges, we present TIGER, a Text-Informed Generalized Enzyme-Reaction Retrieval framework that leverages protein-to-text generation models to distill textual semantic knowledge from enzyme sequences, providing a generalized representation that bridges enzymes and biochemical reactions. To ensure the quality and reliability of textual semantics, we design a Dynamic Gating Network that adaptively fuses text-derived knowledge with sequence features, enabling more consistent and informative enzyme representations, while a Structure-Shared Feature Projector aligns enzyme and reaction representations within a unified latent space. Extensive experiments demonstrate that, under bidirectional retrieval supervision, TIGER significantly outperforms state-of-the-art baselines across diverse distributions and exhibits strong robustness and transferability across tasks.
EVM-QuestBench: An Execution-Grounded Benchmark for Natural-Language Transaction Code Generation
PDF ↗Large language models are increasingly applied to various development scenarios. However, in on-chain transaction scenarios, even a minor error can cause irreversible loss for users. Existing evaluations often overlook execution accuracy and safety. We introduce EVM-QuestBench, an execution-grounded benchmark for natural-language transaction-script generation on EVM-compatible chains. The benchmark employs dynamic evaluation: instructions are sampled from template pools, numeric parameters are drawn from predefined intervals, and validators verify outcomes against these instantiated values. EVM-QuestBench contains 107 tasks (62 atomic, 45 composite). Its modular architecture enables rapid task development. The runner executes scripts on a forked EVM chain with snapshot isolation; composite tasks apply step-efficiency decay. We evaluate 20 models with 5 independent rounds each and find large performance gaps, with split scores revealing persistent asymmetry between single-action precision and multi-step workflow completion. Code: https://github.com/OpenEdgeHQ/EVM-quest-bench.
Large Language Models (LLMs) have shown remarkable capabilities in automating code generation. Recent approaches that incorporate feedback refinement mechanisms into the generation process have further enhanced software generation quality. However, these methods can be characterized as single-path approaches, which suffer from insufficient exploration of the vast solution space, often causing even the most powerful models to get stuck in local optima and struggle to generate the desired software. Some other works use Monte Carlo Tree Search (MCTS) to explore multiple paths for finding the best solution; yet, MCTS can be extremely inefficient in practice. To this end, we propose SeDev, a novel LLM-driven code generation framework that efficiently finds high-quality solutions in only a few iterations. The core idea of SeDev is to gradually explore semantically adjacent solutions through structured prompt guidance and feedback on previous trials, while using unit tests to evaluate the quality of exploration. To distill the exploration experience, SeDev incorporates a feedback synthesis module that translates unit test results within exploration into comprehensive suggestions. We construct a challenging feature oriented software benchmark FSD-bench++, along with two open datasets to evaluate. Experimental results show that SeDev outperforms baselines while maintaining reasonable time and computational costs. Code is available here.
Deploying large language models (LLMs) in resource-constrained environments is hindered by heavy computational and memory requirements. We present LBLLM, a lightweight binarization framework that achieves effective W(1+1)A4 quantization through a novel three-stage quantization strategy. The framework proceeds as follows: (1) initialize a high-quality quantized model via PTQ; (2) quantize binarized weights, group-wise bitmaps, and quantization parameters through layer-wise distillation while keeping activations in full precision; and (3) training learnable activation quantization factors to dynamically quantize activations to 4 bits. This decoupled design mitigates interference between weight and activation quantization, yielding greater training stability and better inference accuracy. LBLLM, trained only using 0.016B tokens with a single GPU, surpasses existing state-of-the-art binarization methods on W2A4 quantization settings across tasks of language modeling, commonsense QA, and language understanding. These results demonstrate that extreme low-bit quantization of LLMs can be both practical and highly effective without introducing any extra high-precision channels or rotational matrices commonly used in recent PTQ-based works, offering a promising path toward efficient LLM deployment in resource-limited situations.