1
A Shared Subcircuit Lets LLMs Count Down Across Tasks
大模型内部发现通用倒数机制，助你在多种任务中精准控制输出长度
arXiv:2607.12279v1 Announce Type: cross Abstract: Writing a sentence of exactly twelve words; ending a DNA sequence at the right codon; formatting an …
2
When Does Continual Learning Require Learning
一作质问持续学习何时真正需要「学习」，揭示任务无关场景下模型无需更新即可泛化，引发对学习本质的再思考。
arXiv:2607.07847v1 Announce Type: new Abstract: As large language models (LLMs) become increasingly capable, the next question is how can we enable mo…
3
Unified Context Evolution for LLM Agents
破解LLM智能体多轮交互中策略遗忘难题，提出统一上下文演化框架，提升跨任务学习效率。
arXiv:2606.02304v1 Announce Type: new Abstract: LLM-based agents can solve multi-step interactive tasks by combining reasoning with environment feedba…