Mapping behavioral intelligence in malware analysis reports to ATT&CK techniques supports threat analysis and defense, but reliable identification remains challenging. Supporting evidence is often scattered across long, heterogeneous reports, while closely related techniques share behavioral cues and differ only in subtle but decisive details. While large language models (LLMs) can use external knowledge to interpret reports, they may assign techniques based on plausible associations rather than sufficient evidence. We present MEvince, a knowledge-constrained LLM reasoning framework for report-level malware technique identification. Its core is a shared behavioral representation connecting report observations to technique definitions. This representation is built through agentic mining of technique documentation and supports two online stages: 1) primitive-guided extraction uses behavioral primitives to structure LLM-extracted observations, link them to source passages, and group dispersed evidence; 2) evidence-aligned retrieval matches this evidence to technique profiles through the same concepts, retaining likely candidates and confusable alternatives for LLM comparison of distinguishing behaviors. Controls at both stages validate evidence references and check technique-specific requirements. The representation also supports adaptation without parameter updates, where offline agentic audits diagnose concept weaknesses and refine extraction or retrieval policies. We evaluate MEvince against 11 baselines spanning four method families using a dataset of 610 public malware reports referenced by MITRE ATT&CK. MEvince improves exact-ID F1 over the strongest baseline by 17.4%, and its concept-guided retrieval increases gold-label candidate coverage from 31.6% to 47.9%. With a small adaptation set, offline refinement yields further gains and outperforms baselines that update model parameters.
Li Wang, Yi Su, Xiabao Wu, Chiran You, Yongchao Liu, Zhan Qiu, Juelu Zhang, Jiajun Zheng 等 12 位作者
Emerging LLMs increasingly interleave full attention with recurrent linear attention to reduce long-context serving costs. Tree speculative decoding could further accelerate these hybrid models, but recurrent linear-attention layers carry parent-dependent states. Existing engines therefore verify proposal trees sequentially and snapshot the full state per node, causing latency and memory to grow with tree and batch sizes. We present Twig, a kernel--runtime co-design for tree speculative decoding of hybrid LLMs. Twig derives a tree-structured factorization of the linear-attention recurrence and realizes it with a resource-efficient GPU kernel that verifies all nodes in parallel, accelerating linear-attention tree verification by 3.4--7.7$\times$. It losslessly represents speculative state updates as token-level factors and reconstructs only the accepted state, reducing transient state memory by 82--99$\times$ and freeing GPU memory for KV caches. We integrate Twig into SGLang, a widely deployed production LLM serving engine, and further optimize its runtime through batch-wide verification budgeting. Across four models, two GPU platforms, and diverse datasets, Twig achieves up to $4.72\times$ the offline decode throughput of autoregressive decoding and up to $2.03\times$ that of the strongest tree-speculative baseline. Under online agent workloads, Twig reduces TTFT and TPOT by up to $67.6\%$ and $49.9\%$, respectively, over the strongest tree-speculative baseline.
Hyper-Connections (HC) architectures are increasingly adopted in frontier language models, yet their components remain insufficiently understood and introduce non-negligible training overhead. In this work, we systematically study the HC architecture to understand its components and identify simpler, more efficient designs. We find that learned read/write gating without mixing and unconstrained mixing with fixed gates form two effective interaction groups: either alone achieves performance comparable to mHC. We also identify repeated access to the expanded residual state as a major source of training overhead, even without mixing. These findings lead to SimpleHC, which simplifies the architecture by removing residual mixing and simplifies execution through shared memory access, using the resulting additive structure and shared gate inputs. Layer-specific gates share a group-entry state, enabling batched reads and deferred writes; scalar corrections exactly preserve preceding sublayer output contributions. Our analyses support both the retained residual structure and the design of shared memory access. SimpleHC incurs only 2.4% extra end-to-end training time over a vanilla residual baseline at the largest tested scale. It maintains performance comparable to mHC when scaling from 15B to 30B MoE models, exhibits smoother gradient dynamics during standard training, and remains stable under stress. We hope SimpleHC will encourage a broader pursuit of simplicity in foundation model design, making performance gains more accessible and practical at scale.
Renáta Rusnáková, Eugen Šlapak, Gaurav K. Pandey, Juraj Gazda
Quantum reservoir computing (QRC) processes temporal information through quantum dynamics, but it remains unclear whether small QRC models can forecast real data competitively with classical models and whether their predictions survive the transfer from ideal simulation to noisy hardware. We address both questions for a six-qubit gate-based reservoir that forecasts next-day motor-insurance claim counts. Five reservoir candidates were selected and frozen using hardware-native interaction graphs, Trotterized Ising dynamics, memory and measurement-robustness criteria, circuit cost, and calibrated-noise screening. On 232 held-out test days, the ideal candidates reached root-mean-square errors (RMSEs) of 5.66-6.12 claims/day, compared with 5.36 for ridge regression. For four of the five, the difference from ridge was not significant under a 95% block bootstrap, and their accuracy was comparable with classical recurrent and reservoir baselines. On the 156-qubit IBM Kingston processor, we ran 63 evaluations in 21 paired rounds, comparing hardware-aware, random-native, and Qiskit optimization level 1 layouts. Hardware-aware placement gave lower RMSE than Qiskit in every round, with a maximum transfer distortion of 7.59 claims/day (11.84 for Qiskit), and avoided the severe transfer failures seen for some random layouts. Most of the hardware distortion appeared as a systematic forecast shift, which a short hardware probe on training inputs detected before the test runs. Longer replay windows raised circuit cost without a consistent gain in forecasting accuracy or hardware transfer. Small QRC models can thus reach competitive accuracy on this task, and hardware-aware layout selection makes their transfer to IBM Kingston more consistent.
Sharyu Kekane, Sadique Khatib, Abhushan Bokade, Uday Wad, S. M. Kamalapur, Y. D. Bhise
Most quantum circuit platforms focus on idealized circuit assembly and simulation, hiding hardware compilation, error identification, and variational model training behind black-box operations. QYANTRAM addresses this gap through an open platform comprising four subsystems, built around evidence-backed results and independent verification. Its staged compilation pipeline makes every transformation traceable, while the router publishes rejected SWAP candidates and their scores for transparent decision-making. Verification exploits inactive qubits remaining in the |0> state, reducing complexity from 2^N to 2^k for k active qubits. This mathematically exact reduction enables full pre-compilation verification on 127- and 156-qubit layouts, achieving approximately 10,000x speed-ups with execution times below one millisecond. The error identification engine exhaustively evaluates all Pauli error patterns without sampling, matching closed-form predictions within 2.8e-17. Its machine-learning engine achieves numerical agreement within 2.0e-15 of qiskit.quantum_info while training 25-50 times faster. The router introduces approximately 1.27x more SWAP gates than Qiskit's SABRE due to its omission of optimization, trading routing efficiency for complete decision transparency and traceability.
Scaling depth is a key route to stronger large language models, but Pre-LN architectures often suffer from depth-wise information dilution, limiting the effective use of deep layers. While cross-layer connectivity has proven effective at mitigating this problem, existing designs usually introduce additional computation or memory access. We revisit cross-layer KV sharing as an efficient form of such connectivity. Beyond cache compression, each KV reuse edge creates a direct gradient path from deep layers to source KV projections, turning KV sharing into an implicit structural regularizer for KV representation geometry. We provide a systematic analysis of this mechanism, showing that proper sharing topologies sharpen key retrieval directions while enriching the value content subspace, whereas existing adjacent or global-anchor designs do not fully exploit this effect. Motivated by this, we propose DepthGraft, a hierarchical KV-sharing architecture that reconstructs deep-layer KV from stride-aligned shallow-middle source pairs. DepthGraft distributes non-local gradient flow across source layers, while preserving prefill early completion and reducing KV cache by roughly one third. Experiments on dense and MoE models from 0.5B to 16B parameters show consistent downstream improvements under both GQA and MLA. Notably, on the 16B MoE model, DepthGraft achieves a 7.0% absolute improvement on MMLU-Pro, along with gains of 6.6% on CMMLU and 6.4% on C3. These results validate the potential of DepthGraft as a scalable design for large-scale pretraining. The project page is available at https://github.com/jianke0604/DepthGraft.
Modern text-to-image generative models increasingly rely on Reinforcement Learning from Human Feedback (RLHF) to align generated content with human preferences and safety standards. This post-training stage is critical for optimizing models toward high-fidelity and value-aligned behaviors using human preference data. However, while previous works have extensively investigated poisoning attacks against the supervised fine-tuning (SFT) stage of text-to-image training, the security vulnerabilities introduced by the RLHF process remain unexplored. In this paper, we present the first exploration into the feasibility of poisoning the RLHF process as a novel attack surface for compromising text-to-image models. We introduce BadReward, an indirect clean-label poisoning attack that manipulates reward signals to embed malicious behaviors. Our approach leverages a feature collision mechanism to produce poison preference image pairs that are visually benign, but aligned with a malicious target concept in the reward feature space. Our experiments demonstrate that this manipulation effectively corrupts the reward signal. Online RLHF further amplifies this corruption through its reward-model feedback loop, whereas offline preference optimization learns directly from the poisoned preferences. Finally, we demonstrate the real-world effectiveness of BadReward on modern text-to-image architectures, showing high attack success rates and visual stealthiness across different RLHF algorithms. Our code is available at https://github.com/ZJUICSR/BadReward.
Olivia Schutz, Rudrajit Choudhuri, Yunhan Qiao, Fatima A. Moussaoui, Julia Russo, Zoe Steine-Hanson, Anita Sarma
Background: Much of the enthusiasm for vibe coding rests on two intuitive beliefs: (1) that developers stay invested even when they no longer write the code, and (2) that delegating more of the work to AI lowers their cognitive load. Objective: Our goal was to investigate whether developers' neurophysiological states and interactions with AI are consistent with these beliefs when they work with agentic vs. non-agentic AI on their own project vs. a legacy codebase. Method: We conducted an exploratory within-subjects crossover study with 11 participants across 22 sessions. Each participant used GitHub Copilot Chat in one session and Cursor Agents in the other, working on their own project and a legacy codebase, with tool-codebase pairings randomly assigned and session order counterbalanced. We recorded participants' EEG, on-screen activity, and prompts, and analyzed 7,924 EEG data points, 984 activities, and 141 prompts. Results: Our findings contradict both beliefs. With the agent, participants remained indifferent to task outcomes, even on their own projects, and cognitive load was higher than with non-agentic AI. Furthermore, their ways of interacting shifted: with non-agentic AI, participants asked how the code worked and read it to verify changes; with the agent, they described desired app behavior and checked results in the running app. Conclusion: Our study reveals a vibe paradox: delegating to an agent left participants motivationally disinterested, even on their own projects, without relieving cognitive load, while coinciding with increased trust in AI. The community should treat this paradox as a first-class concern if developers and their AI agents are to go far together, and not just fast.
We introduce BEACON--Best-Effort Adaptation for Cross-Domain Co-Training--a theory-guided framework for training generative robot policies with abundant source demonstrations and limited target demonstrations. BEACON casts cross-domain co-training as a discrepancy-aware importance-reweighting problem, jointly learning a diffusion-based visuomotor policy and per-sample source weights through an objective inspired by the Best-Effort Adaptation (BEA) bound. To make this approach practical for high-dimensional sequence policies, we develop scalable instance-level discrepancy proxies, stochastic alternating updates for policy and weights, and a multi-source extension that balances heterogeneous source domains. Across sim-to-sim, sim-to-real, and multi-source manipulation settings, BEACON improves robustness and data efficiency over target-only, fixed-ratio co-training, and feature-alignment baselines. Without an explicit alignment objective, BEACON also brings source and target representations closer together as an implicit result of discrepancy-aware co-training.
Rigging is inherently task-dependent because the same mesh may require different skeletons and deformation behaviors across animation tasks. In practice, artists often inspect an initial rig and repeatedly edit its skeletal structure and deformation behavior to meet specific animation requirements. Existing automatic methods primarily generate a plausible rig from geometry, offering limited explicit control over the resulting skeleton and deformation behavior. In this work, we present ViP-Rig, a visual-prompted framework that supports both prompt-first rigging and result-guided editing by injecting features extracted from user-drawn or edited 2D skeletal and rigidity prompts into frozen pretrained backbones. Specifically, ViP-Rig consists of two stages, Skeleton Generation and Skinning Prediction. In the first stage, the skeletal sketch is processed by the Dense-to-Compact Visual Prompt Encoding to produce compact, fixed-length conditioning tokens. The resulting tokens are injected into a frozen pretrained autoregressive generator through gated adapters to control joint placement and branching structure while preserving the generator's geometric prior. In the second stage, the rigidity map is processed using the same visual encoding design, while the pretrained skinning backbone remains frozen. The resulting tokens are symmetrically injected into the point and joint streams to modulate point-joint compatibility and the resulting skinning weights. Experiments on Articulation-XL2.0 and zero-shot evaluation on ModelsResource show that ViP-Rig more accurately recovers target skeletons and skinning weights than geometry-conditioned baselines under prompt-guided evaluation. Qualitative results further demonstrate explicit and localized control in both prompt-first rigging and result-guided editing.
Ptychography is a computational imaging method that reconstructs a sample from overlapping diffraction measurements. This technique relies on these overlapping measurements to provide the redundancy information needed for robust reconstruction, but dense scanning increases acquisition time and data volume. Reducing scan overlaps can improve measurement efficiency but weakens the physical constraints available for reconstruction. Diffusion models provide strong learned prior information, offering a promising approach to compensate for the loss of measurement redundancy. However, effectively utilizing this prior information under weak measurement constraints remains a challenge. We propose a progressive coupling posterior sampling (PCPS) method, which allows for a broader exploration of the solution space under high uncertainty while maintaining stronger diffusion-guided consistency as the reconstruction becomes more refined. Experiments on real X-ray ptychographic data at various scan overlaps demonstrate that PCPS consistently improves reconstruction accuracy compared to traditional iterative methods and existing diffusion posterior samplers, especially at low overlaps. These results show that diffusion priors can reduce the measurement redundancy required for accurate ptychography reconstruction.
Deformable image registration relies on accurate anatomical correspondences. However, explicitly coupling local matching reliability with spatial context selection and deformation update strength remains insufficiently explored in learning-based registration. This challenge is particularly pronounced in homogeneous, weakly textured, and severely deformed regions, where ambiguous matches may induce erroneous deformation updates. We propose ReliMorph, a reliability-routed symmetric framework for 3D deformable registration that couples correspondence assessment with spatial context selection and update strength.Local correlation distributions characterize correspondence ambiguity using shared anatomical features with modality-specific adaptation. These reliability cues guide spatial expert fusion and gate residual velocities during coarse-to-fine refinement. Updated inverse-inconsistency feedback guides routing and gating at finer pyramid levels. To our knowledge, ReliMorph is the first 3D deformable registration framework to jointly condition spatial expert routing and residual velocity gating on correspondence reliability. Extensive experiments demonstrate that ReliMorph consistently outperforms competing methods in anatomical overlap, highlighting its effectiveness across unimodal and multimodal registration tasks.
Shuai Wang, Zihan Qian, Ke Zhang, Jiangyu Han, Zikai Liu, Xiaoyang Yu, Haoyu Li, Marc Delcroix 等 12 位作者
We introduce the REAL-TSE Challenge, an IEEE SLT 2026 satellite challenge on target speaker extraction~(TSE) from real conversational recordings. Given a multi-speaker mixture and one or more enrollment utterances from a target speaker, participating systems must recover only the target speech. Unlike simulated read-speech benchmarks, REAL-TSE evaluates Mandarin and English recordings that contain natural overlap, reverberation, noise, channel mismatch, and conversational dynamics. The challenge defines two complementary tracks: an Online track for low-latency streaming extraction and an Offline track for full-context processing. Systems are evaluated with Token Error Rate (TER), Speaker Similarity (SpkSim), DNSMOS, and target-speaker activity F1. This overview paper describes the task definition, datasets, baselines, evaluation protocol, submitted systems, condition-wise findings, and lessons for future real-world TSE benchmarks.
As quantum devices become accelerators within classical HPC deployments, the efficiency of the classical partition becomes critical. Workloads such as Hamiltonian manipulation, observable backpropagation, and Pauli-based compilation reduce to bulk arithmetic on Pauli strings and sums. We present HPC strategies for this arithmetic, demonstrated in PauLIB, a header-only C++20 library: a bit-packed binary symplectic representation reducing Pauli multiplication to a bitwise XOR and population count; a sorted, struct-of-arrays layout replacing hash maps and exposing contiguous word arrays to SIMD units and GPU warps; and deterministic OpenMP parallelism with race-free output partitioning. These strategies deliver orders-of-magnitude higher throughput and smaller memory footprint than Qiskit and PauliEngine, plus a speedup from a CUDA outer-product kernel on an H100. Two case studies, g-sim and truncated Pauli propagation, inherit these gains and scale unchanged from a desktop Arm CPU to a 72-core HPC Arm server. PauLIB is open source with C++ and Python interfaces.
Chonghan Qin, Xiachong Feng University of Hong Kong, Weitao Ma Harbin Institute of Technology, Xiaocheng Feng, Lingpeng Kong Department of Computer Science, The University of Hong Kong
Existing memory benchmarks for LLM agents evaluate explicit recall of facts, yet overlook implicit memory where experience becomes automated behavior without conscious retrieval. This gap is critical: effective assistants must automatically apply learned procedures or avoid failed actions without explicit reminders. We introduce ImplicitMemBench, the first systematic benchmark evaluating implicit memory through three cognitively grounded constructs drawn from standard cognitive-science accounts of non-declarative memory: Procedural Memory (one-shot skill acquisition after interference), Priming (theme-driven bias via paired experimental/control instances), and Classical Conditioning (Conditioned Stimulus–Unconditioned Stimulus (CS–US) associations shaping first decisions). Our 300-item suite employs a unified Learning/Priming-Interfere-Test protocol with first-attempt scoring. Evaluation of 17 models reveals severe limitations: no model exceeds 66% overall, with top performers DeepSeek-R1 (65.3%), Qwen3-32B (64.1%), and GPT-5 (63.0%) far below human baselines. Analysis uncovers dramatic asymmetries (inhibition 17.6% vs. preference 75.0%) and universal bottlenecks requiring architectural innovations beyond parameter scaling. ImplicitMemBench reframes evaluation from “what agents recall” to “what they automatically enact”.
尚无 AI 解读，展示原文摘要
cs.CL
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表
Shiyao Cui, QingLin Zhang, Di Wang, Yida Lu Tsinghua University, Tsinghua University, Zhexin Zhang, Jinhua Gao Institute of Computing Technology, Chinese Academy of Sciences, Jinglin Yang Institute of Information Engineering, Chinese Academy of Sciences, Min He CNCERT/CC 等 10 位作者
Neologisms, emerging terms in meaning or form, can serve as new vehicles for toxic expression, like “田园女” (“country girl”) as a stigmatizing label targeting feminism. Such toxic neologisms appear benign but have evolved into toxic usage in public consensus, posing challenges to moderation systems and remaining underexplored. In this paper, we investigate how to detect implicit toxicity expressed via neologisms. We first propose a taxonomy that captures the origins and consensus-verification criteria of toxic neologisms, followed by the construction of a lexicon spanning widely observed risk categories. To capture toxicity grounded in public consensus, we introduce SeTox, a search-augmented framework that enables static large language models (LLMs) to incorporate real-time web context for neologism toxicity detection. Experiments show that SeTox, even with 3B-scale models, outperforms recent large-scale models, demonstrating its scalability to incorporate real-world knowledge for toxic neologism detection. Disclaimer: this paper has offensive contents that may be disturbing to some readers.
尚无 AI 解读，展示原文摘要
cs.CL
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表
Renmiao Chen, Yida Lu Tsinghua University, Tsinghua University, Shiyao Cui, Xuan Ouyang, Victor Shea-Jay Huang, Shumin Zhang, Chengwei Pan Beihang Uinveristy, Han Qiu Tsinghua University 等 9 位作者
As Multimodal Large Language Models (MLLMs) acquire stronger reasoning capabilities to handle complex, multi-image instructions, this advancement may pose new safety risks. We study this problem by introducing MIR-SafetyBench, the first benchmark focused on multi-image reasoning safety, which consists of 2,676 instances across a taxonomy of 9 multi-image relations. Our extensive evaluations on 19 MLLMs reveal a troubling trend: models with more advanced multi-image reasoning can be more vulnerable on MIR-SafetyBench. Beyond attack success rates, we find that many responses labeled as safe are superficial, often driven by misunderstanding or evasive, non-committal replies. We further observe that unsafe generations exhibit lower attention entropy than safe ones on average. This internal signature suggests a possible risk that models may over-focus on task solving while neglecting safety constraints.
尚无 AI 解读，展示原文摘要
cs.CL · cs.CV
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表
Large language models (LLMs) are promising for medical question answering (QA) but remain unreliable in Chinese clinical settings due to hallucinations, weak factual grounding, and difficulty handling clinically complex cases. We propose CAMEC (Complexity-Aware Multi-Expert Collaboration), a framework that combines hierarchical medical adaptation with complexity-aware expert routing for reliable Chinese medical QA. We adopt a three-stage LoRA-based supervised fine-tuning pipeline for domain adaptation, instruction following, and clinical reasoning. At inference, CAMEC routes each query by predicted complexity and selectively recruits three experts: an internal chain-of-thought (CoT) expert, a retrieval-augmented expert over a dense medical vector database, and a knowledge graph (KG) expert over a structured medical knowledge base. An LLM-as-a-Judge module evaluates and critiques expert reports, iteratively refining them into a consensus answer. Experiments on four Chinese medical benchmarks show that CAMEC consistentlyoutperforms strong general and medical LLM baselines, achieving 78.86% (CMExam), 84.15% (MedQA-CN), 78.51% (CMMLU-Med), and 74.40% (CMB-exam), with consistent absolute improvements over the previous state-of-the-artHuatuoGPT-o1-7B across all benchmarks. The complexity-aware router reduces expert invocations and inference cost, making CAMEC both highly effective and computationally efficient.
尚无 AI 解读，展示原文摘要
cs.CL
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表
Large language models (LLMs) have demonstrated better safety performance in high-resource languages than in low-resource languages. We attribute this issue as a mismatch gap between language-agnostic semantic understanding ability and language dominant safety alignment biased toward high-resource languages. Based on above insights, we empirically identify the semantic bottleneck in LLMs: intermediate layers in which the geometry of model representations is governed primarily by shared semantic content rather than language identity. Then, we propose Language-Agnostic Semantic Alignment (LASA), which anchors safety alignment directly in semantic bottlenecks. Experiments show that LASA substantially improves safety across all languages: average attack success rate (ASR) drops from 24.7% to 2.8% on LLaMA-3.1-8B-Instruct and remains within 3–4% across Qwen2.5 and Qwen3 Instruct models (7B–32B). Besides, our analysis and method offer a representation-level perspective on LLM safety, suggesting that safety alignment requires anchoring safety understanding not in surface text, but in the model’s language-agnostic semantic space.
尚无 AI 解读，展示原文摘要
cs.CL
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表
Large Language Models have shown remarkable potential as autonomous agents, but their effectiveness in knowledge-intensive tasks remains limited by passive knowledge utilization. We introduce KARL (Knowledge-Augmented Reinforcement Learning), a framework that enables LLM agents to dynamically explore structured knowledge sources through multi-turn interactions. Unlike existing retrieval-augmented approaches, KARL empowers agents to proactively decide when and what knowledge to acquire during task execution. Our framework incorporates online reinforcement learning with curiosity-driven reward shaping, explicitly incentivizing knowledge exploration while optimizing tool-use behaviors end-to-end. Extensive evaluation across six structured knowledge benchmarks demonstrates that KARL achieves state-of-the-art performance, with our Qwen2.5-14B-based agent significantly outperforming GPT-4o, Claude-4, and o4-mini on both knowledge graph and database tasks.Source code is available at https://github.com/THUDM/KARL.
尚无 AI 解读，展示原文摘要
cs.CL · cs.AI
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表
Jianshuo Dong Tsinghua University, Tsinghua University, Yutong Zhang, Liu Yan, Zhenyu Zhong, Tao Wei Ant Group, Chao Zhang Tsinghua University, Han Qiu Tsinghua University
Advanced LLMs have achieved near-ceiling instruction-following accuracy on benchmarks such as IFEval. However, these impressive scores do not necessarily translate to reliable services in real-world use, where users often vary their phrasing, contextual framing, and task formulations. In this paper, we study nuance-oriented reliability: whether models exhibit consistent competence across cousin prompts that convey analogous user intents but with subtle nuances. To quantify this, we introduce a new metric, reliable@k, and develop an automated pipeline that generates high-quality cousin prompts via data augmentation. Building upon this, we construct IFEval++ for systematic evaluation. Across 20 proprietary and 26 open-source LLMs, we find that current models exhibit substantial insufficiency in nuance-oriented reliability—their performance can drop by up to 61.8% with nuanced prompt modifications. What’s more, we characterize it and explore three potential improvement recipes. Our findings highlight nuance-oriented reliability as a crucial yet underexplored next step toward more dependable and trustworthy LLM behavior. Our code and benchmark are accessible: https://github.com/jianshuod/IFEval-pp.
尚无 AI 解读，展示原文摘要
cs.CL
Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · 2026 · 正式发表
An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao 等 60 位作者
In this work, we present Qwen3, the latest version of the Qwen model family. Qwen3 comprises a series of large language models (LLMs) designed to advance performance, efficiency, and multilingual capabilities. The Qwen3 series includes models of both dense and Mixture-of-Expert (MoE) architectures, with parameter scales ranging from 0.6 to 235 billion. A key innovation in Qwen3 is the integration of thinking mode (for complex, multi-step reasoning) and non-thinking mode (for rapid, context-driven responses) into a unified framework. This eliminates the need to switch between different models--such as chat-optimized models (e.g., GPT-4o) and dedicated reasoning models (e.g., QwQ-32B)--and enables dynamic mode switching based on user queries or chat templates. Meanwhile, Qwen3 introduces a thinking budget mechanism, allowing users to allocate computational resources adaptively during inference, thereby balancing latency and performance based on task complexity. Moreover, by leveraging the knowledge from the flagship models, we significantly reduce the computational resources required to build smaller-scale models, while ensuring their highly competitive performance. Empirical evaluations demonstrate that Qwen3 achieves state-of-the-art results across diverse benchmarks, including tasks in code generation, mathematical reasoning, agent tasks, etc., competitive against larger MoE models and proprietary models. Compared to its predecessor Qwen2.5, Qwen3 expands multilingual support from 29 to 119 languages and dialects, enhancing global accessibility through improved cross-lingual understanding and generation capabilities. To facilitate reproducibility and community-driven research and development, all Qwen3 models are publicly accessible under Apache 2.0.