Large Vision-Language Models (LVLMs) enable sophisticated reasoning over images and videos, yet their inference is hindered by a systemic efficiency barrier known as visual token dominance. This overhead is driven by a multi-regime interplay between high-resolution feature extraction, quadratic attention scaling, and memory bandwidth constraints. We present a systematic taxonomy of efficiency techniques structured around the inference lifecycle, consisting of encoding, prefilling, and decoding. Unlike prior reviews focused on isolated optimizations, we analyze the end-to-end pipeline to reveal how upstream decisions dictate downstream bottlenecks, covering compute-bound visual encoding, the intensive prefilling of massive contexts, and the ”visual memory wall” in bandwidth-bound decoding. By decoupling the efficiency landscape into the axes of shaping information density, managing long-context attention, and overcoming memory limits, this work provides a structured analysis of how isolated optimizations compose to navigate the trade-off between visual fidelity and system efficiency. The survey concludes by outlining four future frontiers supported by pilot empirical insights, including hybrid compression based on functional unit sensitivity, modality-aware decoding with relaxed verification, progressive state management for streaming continuity, and stage-disaggregated serving through hardware-algorithm co-design. The submitted software contains a snapshot of our literature repository, which is designed to be maintained as a living resource for the community.
论文检索
输入标题、作者或关键词，从 100,903 篇学术成果中精准定位
Multimodal Large Language Models (MLLMs) rely on strong linguistic reasoning inherited from their base language models. However, multimodal instruction fine-tuning paradoxically degrades this text’s reasoning capability, undermining multimodal performance. To address this issue, we propose a training-free framework to mitigate this degradation. Through layer-wise vision token masking, we reveal a common three-stage pattern in multimodal large language models: early-modal separation, mid-modal alignment, and late-modal degradation. By analyzing the behavior of MLLMs at different stages, we propose a plateau-guided model merging method that selectively injects base language model parameters into MLLMs. Experimental results based on five MLLMs on nine benchmarks demonstrate the effectiveness of our method. Attention-based analysis further reveals that merging shifts attention from diffuse, scattered patterns to focused localization on task-relevant visual regions.Our repository is on https://github.com/wzj1718/PlaM .
Reliable evaluation is essential for understanding large language model (LLM) performance, yet today’s go-to metrics, namely token-overlap scores (e.g., ROUGE) and embedding-based measures (e.g., BERTScore), often misjudge semantic similarity of documents. Our study shows that both token-overlap metrics and embedding-based metrics routinely assign nearly identical scores to texts that directly contradict each other, thereby potentially masking fundamental errors. We introduce MATCHA, an automatic metric that jointly rewards semantic agreement with a reference and penalizes contradictions. MATCHA employs a dual-view perspective that measures (i) proximity to the gold text and (ii) distance from an adversarially generated counterfactual contradiction. In eight public benchmarks, MATCHA outperforms popular metrics, compared with human annotations on question-answering, image caption generation, natural language inference, summarization, and semantic textual similarity tasks. On the TruthfulQA dataset (i.e., a dataset without a training set, where no embedding-based metrics could locally train on), this improvement in terms of matching texts with a reference reaches 18.38% over ROUGE-L and 20.82% over BERTScore. Both quantitative comparison and qualitative human assessments confirm the efficacy and validity of MATCHA and uncover fundamental weaknesses in pre-existing metrics. Compared with 23 embedding models, including top state-of-the-art ones, used as a metric similar to BERTScore, MATCHA remains the most accurate in distinguishing correct from incorrect statements solely based on a reference. Our code and metric are publicly available (https://github.com/Siran-Li/MATCHA).
Although Large Language Models (LLMs) have achieved remarkable success in Machine Translation (MT), a significant performance gap persists between high-resource and low-resource languages due to imbalanced pre-training data. In this paper, we first investigate the internal mechanisms driving this performance disparity from a layer-wise perspective.We propose a metric termed Activation Disparity (\Delta R) to quantify the activation divergence between high- and low-resource MT. Based on this metric, we distinguish between Task-Adaptive Layers (TAL, \Delta R > 0) that encode task-specific signals and Legacy-Inert Layers (LIL, \Delta R < 0) dominated by pre-trained bias. Leveraging this finding, we propose the Layer-aware Dual-directional Modulation (LaDM). Integrated with Low-Rank Adaptation (LoRA), LaDM employs a sparse strategy to bidirectionally modulate optimization dynamics. Specifically, it amplifies contributions from TAL to accelerate feature consolidation while inhibiting LIL to dampen misaligned legacy biases. Extensive experiments on Chinese-to-seven low-resource language translation using Llama-3.1, Qwen2.5, and Gemma-2 demonstrate that LaDM significantly outperforms standard LoRA fine-tuning, achieving an average improvement of 1.73 spBLEU.Code is available at https://github.com/zzssqqq/LaDM.
DiffER: Diffusion Entity-Relation Modeling for Reversal Curse in Diffusion Large Language Models
PDF ↗The "reversal curse" refers to the phenomenon where large language models (LLMs) exhibit predominantly unidirectional behavior when processing logically bidirectional relationships. Prior work attributed this to autoregressive training—predicting the next token inherently favors left-to-right information flow over genuine bidirectional knowledge associations. However, we observe that Diffusion LLMs (DLLMs), despite being trained bidirectionally, also suffer from the reversal curse. To investigate the root causes, we conduct systematic experiments on DLLMs and identify three key reasons: 1) entity fragmentation during training, 2) data asymmetry, and 3) missing entity relations. Motivated by the analysis of these reasons, we propose Diffusion Entity-Relation Modeling (DiffER), which addresses the reversal curse through entity-aware training and balanced data construction. Specifically, DiffER introduces whole-entity masking, which mitigates entity fragmentation by predicting complete entities in a single step. DiffER further employs distribution-symmetric and relation-enhanced data construction strategies to alleviate data asymmetry and missing relations. Extensive experiments demonstrate that DiffER effectively alleviates the reversal curse in Diffusion LLMs, offering new perspectives for future research. The code is available at https://github.com/CQU-MM-Intelligent-Lab/DiffER.
Retrieval-augmented generation (RAG) substantially extends the knowledge boundary of large language models. However, it still faces two major challenges when handling complex reasoning tasks: low context utilization and frequent hallucinations. To address these issues, we propose Self-Correcting RAG, a unified framework that reformulates retrieval and generation as constrained optimization and path planning. On the input side, we move beyond traditional greedy retrieval and, for the first time, formalize context selection as a multi-dimensional multiple-choice knapsack problem (MMKP), thereby maximizing information density and removing redundancy under a strict token budget. On the output side, we introduce a natural language inference (NLI)-guided Monte Carlo Tree Search (MCTS) mechanism, which leverages test-time compute to dynamically explore reasoning trajectories and validate the faithfulness of generated answers. Experiments on six open-domain and multi-hop QA datasets demonstrate that our method significantly improves reasoning accuracy on complex queries while effectively reducing hallucinations, outperforming strong existing baselines. Our code is available at https://github.com/xjiacs/Self-Correcting-RAG .
Large Language Models (LLMs) have shown impressive reasoning capabilities in agents for complex interactive environments. However, these agents often suffer from hallucinations and lack grounding, leading to unreliable actions that conflict with real-world constraints. Existing approaches mitigate some issues through implicit imitation or sparse reinforcement learning but rely on fitting data distributions without explicitly understanding environmental constraints, often generating actions that are behaviorally distorted or environmentally impermissible. To address this, we introduce OntoGuard, an ontological framework designed to guard LLM agents by enforcing environmental and behavioral admissibility. These constraints are constructed by extracting knowledge from oracle demonstrations, supplemented with world knowledge inherent in LLMs and general knowledge bases. During inference, OntoGuard functions as an active interceptor, using a graph-based constraint-checking mechanism to reject invalid actions and prompt self-correction before acting. Experiments on both ScienceWorld and VirtualHome demonstrate OntoGuard’s advantage over state-of-the-art methods, validating its ability to enforce physical and behavioral constraints while preventing invalid actions.
Graph-based retrieval-augmented generation (GraphRAG) is effective for knowledge-intensive and multi-hop query tasks; however, many existing methods primarily seed entity-based graphs and rely on implicit semantic relevance propagation. This often (i) under-retrieves when user queries are abstract and semantically sparse at the entity level, and (ii) suffers from brittle multi-hop reasoning, where noisy activations can derail entity-to-entity transitions and corrupt the inferred relation chain, yielding unreliable conclusions. To this end, we propose FlowRAG, a semantic-aware retrieval framework that improves both semantic recall and explicit reasoning. Specifically, FlowRAG constructs a quad-level heterogeneous graph over passages, summaries, sentences, and entities, where summary nodes serve as a coarse semantic hub. At retrieval time, a dual-granularity activation module combines summary–query alignment with sentence-level matching to activate relevant entities under paraphrase and abstraction robustly. We then introduce a frequency-aware weighted flow module that routes relevance through entity–passage links weighted by within-passage term frequency, pruning noisy connections and extracting high-confidence reasoning paths as an explicit logic skeleton for generation. Extensive experiments show that FlowRAG obtains state-of-the-art performance on complex reasoning benchmarks.
Mixture-of-Experts large language models (LLMs) scale efficiently through sparse activation, yet their deployment is fundamentally constrained by the large static parameter footprint of experts. Existing compression approaches either remove entire experts, disrupting routing topology and harming performance, or rely on unstructured weight pruning with limited practical efficiency. To address the limitations, we propose TENP, a structured **T**rapezoidal **E**xpert **N**euron **P**runing framework. Using a few samples, we identify and retain important experts, while applying expert neuron pruning (ENP) to less important experts, preserving model parameters in a trapezoidal pattern from shallow to deep layers. When evaluating expert importance, we jointly consider both the magnitude of the expert output and its ability to change the direction of the input vector. For ENP, we measure each neuron’s projected contribution to the expert output to identify and retain important neurons. We conduct extensive experiments on the Qwen and DeepSeek models. Under a routing expert sparsity of 40% and an average of 63.76% activated expert parameters, the DeepSeek model suffers only a 1-point drop in accuracy compared to the full-parameter model. Moreover, it outperforms the full-parameter model by 10% on code generation tasks.
Autoregressive (AR) decoding in large language models (LLMs) is latency-bounded by strictly sequential token generation.Speculative decoding mitigates this bottleneck by letting a fast drafter propose multi-token candidates that are then verified in parallel by the target model; yet most existing systems still rely on AR drafters, limiting wall-clock gains.We present **DiffuSpec**, which repurposes a *diffusion language model* (DLM) as a *parallel* drafter to generate multi-token proposals in a single forward pass while remaining compatible with standard AR verifiers.However, DLM drafting presents unique challenges: 1) bidirectional conditioning produces a token lattice where locally optimal tokens may fail to form a valid causal sequence; 2) the mechanism requires tuning the draft length, which induces a speed–quality trade-off. To address these issues, we introduce (i) *Causal-consistency Path Search* (CPS) to extract verifier-aligned causal paths from the lattice, and (ii) an *Adaptive Draft-Length* (ADL) controller that adjusts proposal lengths using online acceptance feedback.Across benchmarks, DiffuSpec achieves up to 3\times wall-clock speedup and consistently outperforms strong baselines, demonstrating diffusion-based drafting as a competitive alternative to AR drafters for speculative decoding.
Beyond Benchmarks: A Capability-Based Maturity Model for Systematic AI Integration in Hospitals
PDF ↗Current Large Language Models (LLMs) demonstrate exceptional performance on medical benchmarks. However, models that excel in standardized tests focused on medical knowledge recall are not necessarily effective in real-world healthcare scenarios. This disparity between academic performance and clinical effectiveness stems from existing evaluations focusing overly on knowledge retrieval and QA, while neglecting high-load executive tasks in real clinical workflows. The effective execution of such tasks depends not only on model reasoning but also on the overall digital maturity of the healthcare institution. To address this, we propose a “Capability-Based Hospital AI Maturity Model” framework. This framework establishes a layered maturity system based on capabilities. By categorizing hospital AI capabilities into distinct maturity levels, it provides a clear, stepwise evolutionary path for hospitals, guiding them from foundational infrastructure construction to ubiquitous intelligence. Guided by this framework, we constructed ten representative real-world clinical scenarios as a reference test set and compared the performance of multiple models across benchmarks and real-world scenarios. Preliminary results suggest that, compared to relying solely on academic benchmark scores, this maturity assessment mode—which integrates system governance and scenario constraints—may provide a more valuable basis for AI adoption in medical institutions.
Long-context LLM agents increasingly serve multiple users or personas within a single session, requiring stable identity and knowledge boundaries under frequent switching. We identify a common failure mode, identity drift, where models conflate user-specific states and leak information across roles. On BEAM-SWITCH, a benchmark for controlled multi-user switching, performance consistently degrades as switching intensifies, even when responses remain fluent and locally coherent. We propose MENTOR, a cognitive architecture that mitigates identity drift without fine-tuning. MENTOR uses a Dual-Chain Memory Mechanism: a Global Chain (\mathcal{G}) for long-term event logging and isolated Role Chains (\mathcal{R}_r) as per-role working memories, supported by a semantic Knowledge Graph (\mathcal{K}) that filters and verifies role-admissible information before generation. Across six LLM families, MENTOR improves the overall score (Avg) from 0.46 to 0.75 on average (+0.29 absolute), with substantial gains in identity adherence and knowledge fidelity.
Large language models (LLMs) achieve strong performance on idiom identification benchmarks, yet their robustness to misleading contextual signals remains largely untested. We introduce ID10M-JAM, an adversarial extension of the ID10M dataset designed to jam model understanding by injecting coherent but conflicting context before each target sentence. For every sentence containing a potential idiomatic expression (PIE), we construct variants that deliberately invert contextual expectations: placing literal cues before idiomatic uses and idiomatic cues before literal ones. All variants are validated by human annotators to ensure naturalness and unambiguous interpretation for human readers. ID10M-JAM exposes systematic vulnerabilities in LLMs’ contextual reasoning, pushing idiom identification to its breaking point.
Omni-modal Large Language Models (OLLMs) excel in diverse tasks but struggle with complex emotional reasoning, which requires integrating textual, visual, and acoustic signals. We attribute this limitation to modality collapse, where models over-rely on a dominant modality while neglecting complementary cues. To address this issue, we introduce OmniCoT, a data paradigm that interleaves guided tokens (e.g., [vision], [audio]) into reasoning traces to enforce structured evidence extraction. To further internalize the reasoning behaviors instilled by OmniCoT and facilitate adaptive modality prioritization, we propose Dynamic Modality-Entropy GRPO (DyME-GRPO), which utilizes entropy-based uncertainty estimates over Guided Tokens (GTs) to regulate modality usage, thereby mitigating collapse and informational redundancy. By applying supervised fine-tuning with OmniCoT followed by DyME-GRPO, we develop EmoOmni based on the Qwen2.5-Omni-7B backbone. Extensive experiments demonstrate that EmoOmni achieves state-of-the-art performance on multiple emotion recognition and reasoning benchmarks while preserving the general capabilities of the base model. These findings highlight the potential of our work for omni-modal reasoning across a broader range of complex tasks.
The rapid advancement and widespread adoption of Large Language Models (LLMs) have elevated the need for reliable AI-generated content (AIGC) detection, which remains challenging as models evolve. We introduce AIGC-text-bank, a comprehensive multi-domain dataset with diverse LLM sources and authorship scenarios, and propose REVEAL, a detection framework that generates interpretable reasoning chains before classification. Our approach uses a two-stage training strategy: supervised fine-tuning to establish reasoning capabilities, followed by reinforcement learning to improve accuracy, improve logical consistency, and reduce hallucinations. Extensive experiments show that REVEAL achieves state-of-the-art performance across multiple benchmarks, offering a robust and transparent solution for AIGC detection. The project is open-source at https://aka.ms/reveal
Reinforcement learning (RL) excels in reasoning tasks with verifiable rewards, while its adaptation to machine translation (MT) remains challenging due to the lack of unique reward signals under multiple valid translations. Existing RL approaches for MT face either fixed references in supervised settings or the production of homogeneous references leading to mode collapse in unsupervised settings. Both limitations arise from ignoring entropy dynamics in RL-based MT. The core challenge is leveraging entropy for supervision construction and self-evolution. In this paper, we propose an Entropy-Driven Unsupervised RL for MT. Our framework integrates entropy-guided sampling for exploration, confidence-weighted label generation to transcend majority-voting bias, and uncertainty-aware optimization to prioritize high-entropy tokens. These mechanisms allow reward signals to co-evolve with model proficiency beyond fixed references. Experiments across multiple language pairs show our method outperforms supervised and unsupervised baselines by +0.63 and +2.52 average points, respectively. Our code is available at https://github.com/fortunatekiss/URLMT.
Code large language models (LLMs) have made significant progress in code debugging by directly generating the correct code based on the buggy code snippet. Programming benchmarks, typically consisting of buggy code snippets and their associated test cases, are used to assess the debugging capabilities of LLMs. However, many existing benchmarks primarily focus on Python and are often limited in terms of language diversity (e.g., DebugBench and DebugEval). To advancethe field of multilingual debugging with LLMs, we propose the first massively multilingual debugging benchmark, which includes 3.9K test samples of 20 programming languages and covers the automated program repair (APR) task, the bug localization(BL) task, and the bug identification (BI) task. In addition, we introduce the debugging instruction corpora MdEval-Instruct by injecting bugs into the correct multilingual queries and solutions (xDebugGen). Further, a multilingual debugger xDebugCoder trained on MdEval-Instruct as a strong baseline specifically to handle bugs of a wide range of programming languages (e.g. “Missing Mut” in language Rust and “Misused Macro Definition” in language C). Our extensive experiments on MdEval reveal a notable performance gap between open-source and closed-source LLMs (e.g., GPT and Claudeseries), highlighting huge room for improvement in multilingual code debugging scenarios.
Speculative Decoding (SD) reduces inference latency for Large Language Models (LLMs) by leveraging an efficient draft model to generate candidate tokens, which are subsequently verified by the target model. To enhance acceleration while reducing the LLM usage costs, we propose Serial and Parallel Intertwined Speculative DEcoding (SPIDE) — a novel training-free SD framework that orchestrates dynamic alternation combining serial dynamic drafting with parallel draft verification. We maintain a confidence-acceptance mapping table during the decoding process. In the serial dynamic drafting module, we leverage this table to evaluate the reliability of the draft sequence and adjust draft lengths adaptively. In the parallel draft verification module, we alleviate drafting-termination conflicts that compromise efficiency, and we update the mapping table synchronously. We conduct experimental evaluations on diverse model pairs and text generation tasks to assess the effectiveness of SPIDE. Compared with autoregressive decoding, SPIDE is speeded up by 3.25× on average and up to 4.56×. Compared with vanilla SD, SPIDE only increases the LLM usage cost by 8.2% on average, but brings an additional 67.7% speedup on average.
Despite advances in safety alignment, Large Language Models (LLMs) remain vulnerable to jailbreaking attacks. However, prevailing methods suffer from a dichotomy of limitations: they either rely on prohibitive iterative optimization in the input space (leading to high computational costs) or fail to penetrate the model’s internal decision-making processes. In this work, we identify a critical structural vulnerability: the ”Attention Sink” mechanism—originally designed to maintain generation stability by anchoring to initial tokens—unintentionally serves as a computational anchor for safety alignment. We hypothesize and empirically verify that safety guardrails are not globally distributed but are predominantly ”front-loaded” in specific attention heads of shallow layers. To audit this structural fragility, we propose FLASH (Focused Layer Attention Sink Hijacking), a novel diagnostic auditing framework that executes a surgical intervention. By precisely scaling attention scores in these vulnerable layers, we dismantle the model’s internal safety anchor. To ensure the attack’s robustness and coherence against the resulting internal noise, we synergize this intervention with multi-variant query rewriting and an adaptive dynamic decoding strategy. Extensive experiments on Llama-3, Qwen-3, and others demonstrate that FLASH achieves a state-of-the-art Attack Success Rate of over 77% with an unprecedented efficiency of 1.53 queries on average. This work marks a paradigm shift from brute-force optimization to mechanism-driven diagnostic auditing, exposing a fundamental trade-off between architectural stability and safety security.
Don’t Just Listen, Try Planning: Graph-based Retrieval-Generation Agent for Long-form Audio Meeting Understanding
PDF ↗Long-form audio meeting understanding (LAMU) is gaining attention, but dedicated question answering (QA) datasets are lacking. Previous tailored speech QA and existing Speech LLMs suffer from acoustic information loss and poor long-term dependency capture. We construct the LongAudioQA dataset and propose the GRGA model, which models heterogeneous audio features into a multi-dimensional graph and leverages agent planning for retrieval and answer generation, effectively addressing existing limitations.