Results across this thread

2 reported results from the papers in this thread.

How this thread developed

1. July 2026 · Alibaba, The Chinese University of Hong Kong

   Self-play with evolving skills boosts LLM benchmarks

   Proposes Skill Self-Play for dynamic skill routing to improve LLM benchmark performance.

2. 1 further paper

   Aug 2026 · Appier AI Research, National Taiwan University

   Joint training helps small language models create and use tools

   Joint training on tool creation and use enables small models to generate their own reusable tools, expanding the skill selection paradigm.

3. 7 further papers

   Sept 2026 · National Key Laboratory for Novel Software Technology, Nanjing University

   Attribution-guided skill graphs improve targeted repairs for frozen language models

   Introduces SkillAA, which builds attribution-guided skill graphs to improve targeted repairs of frozen models' procedural skills without weight updates.

4. Sept 2026 · City University of Hong Kong, National University of Singapore

   Graph-structured skills improve LLM agents through evolutionary optimization

   Graph-structured skills evolved via population-based optimization improve LLM agent performance and context efficiency.

   released code

5. Sept 2026 · East China Normal University, Shanghai AI Laboratory

   Verified workflow training gives language models reusable procedural skills

   Creates executable environments from human-written skills and collects trajectories to train models on reusable procedural skills.

6. Sept 2026 · Lingnan University, The Hong Kong Polytechnic University

   Reusable skills improve smart-contract audits, but model choice dominates

   Evaluates how wild-collected skills affect smart-contract audit agent performance, showing model choice dominates over skill selection.

6 of 14 papers shown