📄 arXiv 论文速递

💡 针对难以用规则实现但描述简单的文本函数，提出将自然语言规范训练为本地神经函数的方法，避免每次调用远程大模型，兼顾效率与灵活性，对实际应用有重要价值。

Many recurring text functions are easy to describe but difficult to implement with rules, while calling a large remote model for every input introduces repeated cost, latency, and …

💡 揭示黑盒LLM作为评测工具在共享端点上的可靠性缺陷，通过预注册实验证明其测量不稳定，对依赖LLM评判数据质量、模型排序的研究和实践提出关键警示。

Language-model judges now gate training data, score generations, and drive leaderboards. The judge is then a measurement instrument, resting on one rarely stated assumption: the sa…

💡 针对进化式提示优化器存在的提示膨胀问题，提出错误结构化优化方法，通过诊断、多样化和稳定化策略生成更简洁有效的提示，显著提升优化效率与性能。

Evolutionary prompt optimizers such as GEPA suffer from prompt bloat: each iteration appends rules and caveats, producing prompts up to 3$\times$ longer yet no more accurate. We tr…

💡 质疑思维链推理轨迹的可读性等同于可解释性，通过对比人类判断与实际重要性，揭示认知偏差，对理解LLM推理机制及安全评估具有重要理论意义。

Reasoning traces from chain-of-thought models appear to offer a legible window into how a model arrives at its answer. A growing body of work treats them as such, using LLM judges …

Video editing spans diverse editing paradigms, yet achieving high-quality instruction-guided and subject-guided editing within a single unified framework remains challenging. We in…

We introduce the first Probably Approximately Correct (PAC) learning framework for general-sum concurrent stochastic games (CSGs) with transition uncertainty, while addressing the …

Weakly-Supervised Dense Video Captioning aims to localize and describe multiple events in untrimmed videos given only an ordered set of event-level captions per video. Recent work …

Gaps remain in our understanding of how large language models (LLMs) acquire knowledge during pre-training. We posit that auxiliary views, reformulations of knowledge, are causally…

💡 提出计算可行的因果概率解释框架，系统回答特定结果的原因及输入的责任归属，弥合哲学理论与实际应用间的鸿沟，为可解释AI提供新工具。

Explaining why a specific outcome occurred, and which inputs deserve the blame or credit, is central to philosophical, scientific, and policy analysis. Existing tools split into tw…

On-policy distillation (OPD) combines student-generated rollouts with dense token-level supervision from a teacher. Existing work has mainly studied its algorithmic behavior, leavi…

Multi-agent AI science ecosystems rely on agents possessing tools that allow them to communicate, coordinate, and build on each other's work. Yet this shared infrastructure can als…

As edge-based deep learning applications become more complex, optimizing performance on heterogeneous System-on-Chips (SoCs) presents unique challenges. Traditional pipelining tech…