arXiv:2509.22097v2 Announce Type: replace-cross Abstract: Large language model-powered code agents are rapidly transforming software engineering, yet the security risks of their generated code have become a critical concern. Existing benchmarks have provided valuable insights, but they fail to capture scenarios in which vulnerabilities are actually introduced by human developers, making fair comparisons between humans and agents infeasible. We therefore introduce SecureVibeBench, a benchmark of 105 C/C++ secure coding tasks sourced from 41 projects in OSS-Fuzz for code agents. SecureVibeBench has the following features: (i) realistic task settings that require multi-file edits in large repositories, (ii)~aligned contexts based on real-world open-source vulnerabilities with precisely identified vulnerability introduction points, and (iii) comprehensive evaluation that combines functionality testing and security checking with both static and dynamic oracles. We evaluate 5 popular code agents like OpenHands, supported by 5 LLMs (e.g., Claude sonnet 4.5) on SecureVibeBench. Results show that current agents struggle to produce both correct and secure code, as even the best-performing one, produces merely 23.8\% correct and secure solutions on SecureVibeBench.

arXiv:2510.15994v2 Announce Type: replace-cross Abstract: The Model Context Protocol (MCP) standardizes how large language model (LLM) agents discover, describe, and call external tools. While MCP unlocks broad interoperability, it also enlarges the attack surface by making tools first-class, composable objects with natural-language metadata, and standardized I/O. We present MSB (MCP Security Benchmark), the first end-to-end evaluation suite that systematically measures how well LLM agents resist MCP-specific attacks throughout the full tool-use pipeline: task planning, tool invocation, and response handling. MSB contributes: (1) a taxonomy of 12 attacks including name-collision, preference manipulation, prompt injections embedded in tool descriptions, out-of-scope parameter requests, user-impersonating responses, false-error escalation, tool-transfer, retrieval injection, and mixed attacks; (2) an evaluation harness that executes attacks by running real tools (both benign and malicious) via MCP rather than simulation; and (3) a robustness metric that quantifies the trade-off between security and performance: Net Resilient Performance (NRP). We evaluate nine popular LLM agents across 10 domains and 405 tools, producing 2,000 attack instances. Results reveal the effectiveness of attacks against each stage of MCP. Models with stronger performance are more vulnerable to attacks due to their outstanding tool calling and instruction following capabilities. MSB provides a practical baseline for researchers and practitioners to study, compare, and harden MCP agents. Code: https://github.com/dongsenzhang/MSB

arXiv:2603.12271v1 Announce Type: cross Abstract: LLMs are widely used in knowledge-intensive tasks where the same fact may be revised multiple times within context. Unlike prior work focusing on one-shot updates or single conflicts, multi-update scenarios contain multiple historically valid versions that compete at retrieval, yet remain underexplored. This challenge resembles the AB-AC interference paradigm in cognitive psychology: when the same cue A is successively associated with B and C, the old and new associations compete during retrieval, leading to bias. Inspired by this, we introduce a Dynamic Knowledge Instance (DKI) evaluation framework, modeling multi-updates of the same fact as a cue paired with a sequence of updated values, and assess models via endpoint probing of the earliest (initial) and latest (current) states. Across diverse LLMs, we observe that retrieval bias intensifies as updates increase, earliest-state accuracy stays high while latest-state accuracy drops substantially. Diagnostic analyses of attention, hidden-state similarity, and output logits further reveal that these signals become flatter and weakly discriminative on errors, providing little stable basis for identifying the latest update. Finally, cognitively inspired heuristic intervention strategies yield only modest gains and do not eliminate the bias. Our results reveal a persistent challenge in tracking and following knowledge updates in long contexts.

arXiv:2603.13023v1 Announce Type: cross Abstract: Training capable software engineering (SWE) agents demands large-scale, executable, and verifiable environments that provide dynamic feedback loops for iterative code editing, test execution, and solution refinement. However, existing open-source datasets remain limited in scale and repository diversity, while industrial solutions are opaque with unreleased infrastructure, creating a prohibitive barrier for most academic research groups. We present OpenSWE, the largest fully transparent framework for SWE agent training in Python, comprising 45,320 executable Docker environments spanning over 12.8k repositories, with all Dockerfiles, evaluation scripts, and infrastructure fully open-sourced for reproducibility. OpenSWE is built through a multi-agent synthesis pipeline deployed across a 64-node distributed cluster, automating repository exploration, Dockerfile construction, evaluation script generation, and iterative test analysis. Beyond scale, we propose a quality-centric filtering pipeline that characterizes the inherent difficulty of each environment, filtering out instances that are either unsolvable or insufficiently challenging and retaining only those that maximize learning efficiency. With $891K spent on environment construction and an additional $576K on trajectory sampling and difficulty-aware curation, the entire project represents a total investment of approximately $1.47 million, yielding about 13,000 curated trajectories from roughly 9,000 quality guaranteed environments. Extensive experiments validate OpenSWE's effectiveness: OpenSWE-32B and OpenSWE-72B achieve 62.4% and 66.0% on SWE-bench Verified, establishing SOTA among Qwen2.5 series. Moreover, SWE-focused training yields substantial out-of-domain improvements, including up to 12 points on mathematical reasoning and 5 points on science benchmarks, without degrading factual recall.

arXiv:2603.06665v1 Announce Type: cross Abstract: Large vision-language models (VLMs) often benefit from chain-of-thought (CoT) prompting in general domains, yet its efficacy in medical vision-language tasks remains underexplored. We report a counter-intuitive trend: on medical visual question answering, CoT frequently underperforms direct answering (DirA) across general-purpose and medical-specific models. We attribute this to a \emph{medical perception bottleneck}: subtle, domain-specific cues can weaken visual grounding, and CoT may compound early perceptual uncertainty rather than correct it. To probe this hypothesis, we introduce two training-free, inference-time grounding interventions: (i) \emph{perception anchoring} via region-of-interest cues and (ii) \emph{description grounding} via high-quality textual guidance. Across multiple benchmarks and model families, these interventions improve accuracy, mitigate CoT degradation, and in several settings reverse the CoT--DirA inversion. Our findings suggest that reliable clinical VLMs require robust visual grounding and cross-modal alignment, beyond extending text-driven reasoning chains. Code is available \href{https://github.com/TianYin123/Better_Eyes_Better_Thoughts}{here}.

arXiv:2510.10285v3 Announce Type: replace Abstract: Multimodal large reasoning models (MLRMs) often suffer from hallucinations that stem not only from insufficient visual grounding but also from imbalanced allocation between perception and reasoning processes. Building upon recent interpretability findings suggesting a staged division of attention across layers, we analyze how this functional misalignment leads to two complementary failure modes: perceptual bias in shallow layers and reasoning drift in deeper layers. To alleviate these issues, we propose Functional Head Identification and Class-Conditioned Rescaling , a lightweight, training-free plugin that identifies perception- and reasoning-oriented heads and adaptively rebalances their layerwise contributions. Our method improves reasoning consistency and visual faithfulness without retraining or any architectural modification. Evaluations across three representative MLRMs and five multimodal reasoning benchmarks show an average 4.2% point gain, with less than 1% additional computation and only 9% baseline latency. Beyond empirical improvements, our study provides an interpretable perspective on regulating cross-layer functional dynamics to enhance the reliability of multimodal reasoning.

arXiv:2602.12851v2 Announce Type: replace-cross Abstract: Deploying expressive learning models directly on programmable dataplanes promises line-rate, low-latency traffic analysis but remains hindered by strict hardware constraints and the need for predictable, auditable behavior. Chimera introduces a principled framework that maps attention-oriented neural computations and symbolic constraints onto dataplane primitives, enabling trustworthy inference within the match-action pipeline. Chimera combines a kernelized, linearized attention approximation with a two-layer key-selection hierarchy and a cascade fusion mechanism that enforces hard symbolic guarantees while preserving neural expressivity. The design includes a hardware-aware mapping protocol and a two-timescale update scheme that together permit stable, line-rate operation under realistic dataplane budgets. The paper presents the Chimera architecture, a hardware mapping strategy, and empirical evidence showing that neuro-symbolic attention primitives can achieve high-fidelity inference within the resource envelope of commodity programmable switches.

arXiv:2603.02854v1 Announce Type: cross Abstract: Language-conditioned navigation pipelines often rely on brittle modular components or costly action-sequence generation. To address these limitations, we present CoFL, an end-to-end policy that directly maps a bird's-eye view (BEV) observation and a language instruction to a continuous flow field for navigation. Instead of predicting discrete action tokens or sampling action chunks via iterative denoising, CoFL outputs instantaneous velocities that can be queried at arbitrary 2D projected locations. Trajectories are obtained by numerical integration of the predicted field, producing smooth motion that remains reactive under closed-loop execution. To enable large-scale training, we build a dataset of over 500k BEV image-instruction pairs, each procedurally annotated with a flow field and a trajectory derived from BEV semantic maps built on Matterport3D and ScanNet. By training on a mixed distribution, CoFL significantly outperforms modular Vision-Language Model (VLM)-based planners and generative policy baselines on strictly unseen scenes. Finally, we deploy CoFL zero-shot in real-world experiments with overhead BEV observations across multiple layouts, maintaining reliable closed-loop control and a high success rate.