llms

Documents about llms

• ConQuR: Corner Aligned Activation Quantization via Optimized Rotations for LLMs#175273arXiv CS
• DynaMiCS: Fine-tuning LLMs with Performance Constraints using Dynamic Mixtures#175280arXiv CS
• BED-LLM: intelligent information gathering with LLMs and Bayesian experimental design#183861Ora Oxford University Research
• Understanding Inference Scaling for LLMs: Bottlenecks, Trade-offs, and Performance Principles#204751arXiv CS
• Can LLMs Produce Better Object-Oriented Designs than Human-Involved Development?#204840arXiv CS
• SWE-Mutation: Can LLMs Generate Reliable Test Suites in Software Engineering?#216890arXiv CS
• RefusalBench: Why Refusal Rate Misranks Frontier LLMs on Biological Research Prompts#216910arXiv CS
• DiscoverPhysics: Benchmarking LLMs for Out-of-the-Box Scientific Thinking#224499arXiv CS
• Pair-In, Pair-Out: Latent Multi-Token Prediction for Efficient LLMs#229535arXiv CS
• ProjectionBench: Evaluating Scientific Hypothesis Generation in LLMs Under Progressive Information Disclosure#238649arXiv CS
• CrunchLLM: Multitask LLMs for Structured Business Reasoning and Outcome Prediction.#244236NCBI PubMed Central
• Willing but Unable: Separating Refusal from Capability in Code LLMs via Abliteration#259378arXiv CS
• Empirical Study for Structured Output Control in LLMs for Software Engineering#267711arXiv CS
• Snyk VulnBench JS 1.0: Can LLMs Find the Same Bugs Twice?#280133arXiv CS
• When LLMs Analyze Scars: From Images to Clinically-Meaningful Features#282793arXiv CS
• SoftMoE: Soft Differentiable Routing for Mixture-of-Experts in LLMs#282806arXiv CS
• SAC: Disaggregated KV Cache System for Sparse Attention LLMs with CXL#290553arXiv CS
• What Do Safety-Aligned LLMs Learn From Mixed Compliance Demonstrations?#290566arXiv CS
• Where Do Models Find Happiness? Emotion Vectors in Open-Source LLMs#310866arXiv CS
• Diff-Based Code Corruption using LLMs for Large-Scale Bugfix Benchmarking#321913arXiv CS
• Reinforcement Learning with Metacognitive Feedback Elicits Faithful Uncertainty Expression in LLMs#324919arXiv CS
• When LLMs Read Tables Carelessly: Measuring and Reducing Data Referencing Errors#324920arXiv CS
• Towards Knowledge Alignment in Code LLMs: Contrastive Unlearning for Evolving APIs#324973arXiv CS
• WattGPU: Predicting Inference Power and Latency on Unseen GPUs and LLMs#332487arXiv CS
• PathMark: Protecting Intellectual Property of Mixture-of-Expert LLMs via Path Watermarks#343402arXiv CS
• Teaching LLMs a Low-Resource Language: Enhancing Code Completion in Pharo#343539arXiv CS
• Beyond the Syntax: Do Security Experts Trust LLMs for NIDS Rule Engineering?#346451arXiv CS
• Super Weights in LLMs and the Failure of Selective Training#353041arXiv CS
• The Illusion of Equivalency: Statistical Characterization of Quantization Effects in LLMs#353089arXiv CS
• Using LLMs to Adjudicate Static-Analysis Alerts with Error Reduction Techniques#363350arXiv CS
• DataShield: Uncovering Risky Fine-Tuning Data Across LLMs Through Consensus Subspace Alignment#373322arXiv CS
• Innocuous-Seeming Data, Latent Ideology: Ideological Generalisation in Finetuned LLMs#373400arXiv CS
• Beyond Generalist LLMs: Specialist Agentic Systems for Structured Code Workflow Execution#373464arXiv CS
• NexForge: Scaling Agent Capabilities through Requirement-Driven Task Synthesis for LLMs#381815arXiv CS
• Breaking Customized LLMs for Coding: Automated Red Teaming for Instruction Backdoor Attacks#432464arXiv CS
• It's How You Ask: Gender-Associated Linguistic Bias in LLMs#451346arXiv CS
• <b>TCM-LCR</b>#454392Figshare
• AnchorBench: A Multi-Pathway Benchmark for the Anchoring Effect in LLMs#458335arXiv CS
• Toward Better Assessment of LLMs' Performance in Clinical Error Detection#459665arXiv CS
• OpenHarmony Bench: Evaluating LLMs and Coding Agents on OpenHarmony App Development#459734arXiv CS
• Implementing a large language model from scratch#462355RUcore, Rutgers University Community Repository
• FedLNS: Leverage LayerNorm Signature Modeling to Mitigate Adversarial Manipulation in Federated LLMs#467079arXiv CS
• Structured but Fragile: On the Limits of LLMs in Cybersecurity Decision-Making#479993arXiv CS
• Beyond the Traceback: Using LLMs for Adaptive Explanations of Programming Errors#480126arXiv CS
• RamseyGadgets: A Graph Construction Dataset for LLMs#610957arXiv (OAI Expanded)
• RamseyGadgets: A Graph Construction Dataset for LLMs#612434arXiv (All)
• LLMs Get Smarter from Targeted Synthetic Multilingual Data#616576arXiv (OAI Expanded)
• LLMs Get Smarter from Targeted Synthetic Multilingual Data#617076arXiv (All)
• OpenHarmony Bench: Evaluating LLMs and Coding Agents on OpenHarmony App Development#617195arXiv (OAI Expanded)
• OpenHarmony Bench: Evaluating LLMs and Coding Agents on OpenHarmony App Development#617695arXiv (All)