This paper presents a method for compressing matrices using a RePair straight-line program (SLP), allowing matrix-vector products with time and space proportional to the compressed size, and demonstra…

The paper introduces Kumushi, a root-cause-driven patching agent that significantly improves automated vulnerability repair by focusing LLMs on the true source of bugs, outperforming existing methods…

The paper introduces TEERepair, a framework that automatically repairs severe security vulnerabilities caused by improper partitioning in Trusted Execution Environments (TEEs) by combining a domain-sp…

The paper introduces Semantic Triplet Restoration (STR), a novel protocol that converts complex table structures into atomic semantic triplets, improving table question answering by providing explicit…

The paper introduces LogicEval, a systematic framework and dataset (LogicDS) to evaluate automated repair techniques for logical software vulnerabilities, finding that prompt sensitivity and context l…

This paper proposes a method to recover recoverability structure from failed traces of post-trained language models, enabling test-time routing and post-training analysis.

This paper introduces Re-Sonance, a real-time speech-driven AAC system for professional speaking scenarios using LLM-enhanced Whisper ASR, Qwen LLM, and CosyVoice TTS.

This paper compares signal-triggered and fixed-cadence repair policies for graph approximate-nearest-neighbor indexes and shows that signal-triggered repair improves worst-case recall at scarce budget…

This paper identifies the root cause of performance degradation in full-duplex Spoken Language Models (SLMs) due to modality interference and proposes Lychee-FD, a framework that decouples conflicting…

肖代替了视觉令牌的永久删除，通过可恢复的路由来改进视觉语言模型的性能

Introduces SVD-Surgeon, a training-free method for improving the perplexity-compression trade-off in large language models using SVD, by updating retained singular values and providing a saliency for…

Eight voice cloning models are benchmarked on five paralinguistic tasks, showing most preserve signal with modest degradation. Cloning English clinical speech into Japanese outperforms raw cross-lingu…

This paper presents TerraRepair, a tool-grounded LLM agent for Terraform repair with structured escalation, which improves scanner-verified fix rates for IaC security scanners Checkov and Trivy on AWS…

This paper proposes CoHarden, a co-generation framework for automated program repair that uses a lax signal as an in-loop convergence criterion to prevent lax regressions.

ContraFix is an agentic framework that improves automated vulnerability repair by using differential runtime evidence to pinpoint the root cause of bugs, achieving state-of-the-art performance on majo…

VulKey introduces a novel LLM-based framework that uses a hierarchical abstraction of expert security knowledge to guide automatic vulnerability repair, achieving state-of-the-art performance on real-…

The paper proposes Locality-Aware Redundancy Pruning (LoRP), a training-free method that prunes LLM layers by exploiting localized inter-layer redundancy, leading to improved efficiency while maintain…

The paper introduces TalkTag, an LLM-based tool that automates fine-grained morphosyntactic error annotation for spoken-language transcripts, providing a scalable alternative to labor-intensive manual…

The paper introduces Repair-Augmented Constraint Learning (RACL), a framework that models contextual decisions by allowing systems to learn whether a candidate should be repaired before being vetoed,…