用任务向量让大模型生成既功能正确又安全无漏洞的代码，新方法直击LLM代码生成痛点
arXiv:2607.07881v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly used for code generation, but they struggle to generat…
LLM说“我很有把握”其实更代表“我坚持选它”，而非正确答案——揭秘AI置信度的真实含义。
arXiv:2606.29490v1 Announce Type: new Abstract: Confidence is an estimate of the probability that a chosen answer is correct. Verbal confidence report…
探索LLM隐藏状态如何预判代码正确性，并解析修复过程中的几何信号，为代码生成质量提供新洞察。
arXiv:2606.14530v1 Announce Type: new Abstract: Large language models encode rich information in their hidden states. This work asks whether code corr…
利用不同模型间的分歧作为无标签正确性信号，破解自信错误检测难题。
arXiv:2603.25450v2 Announce Type: replace Abstract: Detecting when a language model is wrong without ground truth labels is a fundamental challenge fo…
将LLM编译成可验证正确性的CUDA megakernel，大幅提升推理性能与可靠性。
Article URL: https://github.com/RightNow-AI/AutoMegaKernel Comments URL: https://news.ycombinator.com/item?id=48447329 Points: 2 # Comments: 0
从追求正确性转向衡量实用性，这一新方法革新了LLM推理过程中的前缀评估策略。
arXiv:2606.07190v1 Announce Type: new Abstract: Reasoning prefixes shape the future trajectory of LLM problem solving, yet existing process reward mod…
从LLM幻觉到人类认知偏差：学习停滞的解剖，揭示正确性与真理的鸿沟
Article URL: https://tagide.com/blog/llm/the-anatomy-of-a-learning-stall/ Comments URL: https://news.ycombinator.com/item?id=48435840 Points: 2 # Comm…
提出超越正确性、奖励推理忠实性的RAG方法，为AI对齐提供新思路。
arXiv:2510.13272v3 Announce Type: replace Abstract: Inspired by the success of reinforcement learning (RL) in Large Language Model (LLM) training for …
LLM训练新思路：从“饱和数据”中挖掘超越正确性的隐藏信号
arXiv:2606.01436v1 Announce Type: new Abstract: The growing capabilities of large language models (LLMs) have led to the saturation of many benchmarks…
Opt-Verifier通过双端验证机制，提升大语言模型构建数学优化模型的正确性，解决自动化建模中的关键难题。
arXiv:2605.29556v1 Announce Type: new Abstract: Building mathematical optimization models is critical in operations research (OR), while it requires s…
新方法从代码规范自动推断正确性，无需运行即可验证，提升代码质量与可靠性。
arXiv:2605.29822v1 Announce Type: cross Abstract: Large language models (LLMs) have become integral to modern software development, enabling automated…
提出功能熵方法，通过不确定性量化预测LLM生成代码的功能正确性，为代码可靠性评估提供全新视角
arXiv:2605.28500v1 Announce Type: cross Abstract: Large language models have shown impressive capabilities in code generation, yet they often produce …
用智能体引导树搜索，让大模型自动生成形式化验证代码，降低软件验证门槛
arXiv:2605.27485v1 Announce Type: cross Abstract: Formal verification offers a path to provably correct software, but writing verified code remains ex…
外推权重平均法揭示了代码强化学习中正确性与效率的权衡前沿，为模型调优提供新视角。
arXiv:2605.28751v1 Announce Type: cross Abstract: Linear interpolation between fine-tuned checkpoints has been shown to trace the Pareto front between…
探索LLM智能体能否从历史失败中学习，解决代码仓库环境配置难题——依赖冲突等常见坑点不再束手无策。
arXiv:2605.26186v1 Announce Type: cross Abstract: Functionality-correct repository setup aims to configure execution environments (e.g., dependencies,…
置信度几何方法揭示大语言模型推理中痕迹级正确性的全新框架，附开源代码。
arXiv:2605.16824v1 Announce Type: new Abstract: Large language models (LLMs) generate not only reasoning text, but also token-level confidence traject…
探讨Chain-of-Thought验证器的在线可学习性，深入分析正确性与完备性间的权衡关系。
arXiv:2603.03538v3 Announce Type: replace Abstract: Large Language Models (LLMs) with chain-of-thought generation have demonstrated great potential fo…
提出Agent技能作为可验证工件，用双条件正确性标准解决人机协作信任问题，LLM部署的新范式
arXiv:2605.00424v2 Announce Type: replace-cross Abstract: Agent skills - structured packages of instructions, scripts, and references that augment a l…