Tag
该论文通过隐藏的PDDL形式语义对LLM智能体的规划能力进行了严格验证，发现最佳配置离线可解78%的任务但在线仅40%，且复杂提示策略在无明显收益的情况下多消耗5至22倍token，质疑了现有LLM智能体规划能力的声明。
This paper introduces μSkia, a formal semantics for the Skia 2D graphics library, and develops an optimizer that achieves 18.7% speedup on real-world Skia programs from top websites, verified using Lean.
This paper develops a vector logic for formal semantics, characterizing when compressed vector representations allow exact linear or affine readouts of truth conditions, with experiments on GloVe and word2vec embeddings.
This paper mathematically models Classical Arabic morphology to argue against Saussurean arbitrary sign and postmodern semantic indeterminacy, proving that Arabic lexical meaning is structurally determined via root-pattern pairs and showing a complexity asymmetry over Indo-European languages.
This paper measures how much formal semantic structure explains human label variation in natural language inference (NLI) using ChaosNLI data, finding group-level effects on entropy but item-level ceilings and null composition effects.
This paper presents a formally grounded ODRL evaluator with transparent semantics, supporting all rule types, and compares its performance with existing evaluators, addressing interoperability issues in policy modelling for data access and AI governance.
This paper presents a formal theory for defining reuse of answers in governed conversational AI systems, replacing similarity heuristics with mathematically characterized quotient spaces of resolved utterances.
A personal research page cataloging small finitely-presented monoids and their word-problem difficulty, motivated by Swift compiler’s use of Knuth-Bendix completion for generic-signature equivalence.