Chen Huang

Jingyao Liu, Danling Meng, Chen Huang et al.

Distilling historical trajectories into reusable experience to enhance future problem-solving has become a focal point of recent LLM research. However, existing methods predominantly operate at the task level, leveraging general summaries or rules under the assumption that analogous tasks share universal solution patterns. This approach often fails in complex reasoning, which typically falters at local bottlenecks that require precise, state-specific guidance rather than broad heuristics. We introduce HippoSpark, a state-level experience system that performs on-demand retrieval tailored to the immediate needs of the current reasoning state. Across mathematical, scientific, and programming benchmarks, HippoSpark consistently outperforms both standard prompting and task-level experience baselines. Our findings reveal that the most effective experience systems are those that provide actionable guidance at critical bottlenecks rather than serving as generic task-level context. Our code is available at https://github.com/DanlingMeng/HippoSpark.

Chen Huang, Xuefeng Wen

This paper presents two logic systems, SCP and its extension SCP1, to distinguish between different types of impossibility. The semantics use a stratified structure that partitions worlds into logic-normal worlds (N) and anti-logic worlds (I). By defining a metaphysical accessibility relation within N, the systems separate metaphysical impossibility from logical contradiction. This allows for logical reasoning to be maintained even when dealing with metaphysical impossibilities. To address the challenge of vacuism, SCP1 implements a non-empty constraint on the selection function, ensuring that counterpossibles with impossible antecedents are not trivially true but depend on the connection between the antecedent and the consequent. We provide proofs for the soundness, completeness, and decidability of both systems. Finally, we indicate the possibility of applying this stratified approach to other modal domains, such as deontic or epistemic logic.