analiza aiqbench
SALA introduces a novel framework for in-context learning that shifts from rigid, surface-level retrieval to flexible semantic alignment of reasoning operations. By embedding task-specific logic into a continuous space and utilizing dynamic time warping, the method effectively handles diverse problem-solving processes while maintaining interpretability. Experimental results across multiple benchmarks confirm its superiority over existing demonstration selection techniques for complex reasoning tasks.
SALA: Semantic-Aware Logical Alignment for Complex Reasoning in In-Context Learning
arXiv:2609.02336v1 Announce Type: cross Abstract: Effective in-context learning (ICL) for complex reasoning relies on selecting the right demonstrations. Traditional retrieval methods based on surface similarity fail to capture the underlying problem-solving logic. Recent logic-based methods address this by matching predefined reasoning steps, but the rigid rules and exact-match criteria is improper to handle flexible or diverse reasoning processes. To address the problem, we propose SALA, a Semantic-Aware Logical Alignment framework. Instead of relying on a fixed inventory, SALA automatically learns task-specific reasoning operations. It then embeds these operations into a continuous semantic space and uses dynamic time warping (DTW) to align the reasoning sequences. This approach allows for soft, flexible matching of reasoning logic while remaining highly interpretable. Experiments across four reasoning benchmarks and three LLMs demonstrate that SALA outperforms existing demonstration selection methods. Further analysis confirms the roles of the operation induction and the logical semantic alignment.
Abstract:Effective in-context learning (ICL) for complex reasoning relies on selecting the right demonstrations. Traditional retrieval methods based on surface similarity fail to capture the underlying problem-solving logic. Recent logic-based methods address this by matching predefined reasoning steps, but the rigid rules and exact-match criteria is improper to handle flexible or diverse reasoning processes. To address the problem, we propose SALA, a Semantic-Aware Logical Alignment framework. Instead of relying on a fixed inventory, SALA automatically learns task-specific reasoning operations. It then embeds these operations into a continuous semantic space and uses dynamic time warping (DTW) to align the reasoning sequences. This approach allows for soft, flexible matching of reasoning logic while remaining highly interpretable. Experiments across four reasoning benchmarks and three LLMs demonstrate that SALA outperforms existing demonstration selection methods. Further analysis confirms the roles of the operation induction and the logical semantic alignment.
Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
arXiv:2609.02336 [cs.AI]
(or arXiv:2609.02336v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.02336 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 241 cuvinte
Subiecte conexe
Research Note: Filtering Subversion-Relevant Information From Pretraining Data Is Feasible
LessWrong · 5 oct. 2026
[AINews] Meta Muse Wonder & Wonder Full: MSL’s Claude 5.5-tier Frontier Model
Latent Space · 13 oct. 2026
Game Design Creativity evals
LessWrong · 12 oct. 2026
FlowRAG: Synergizing Explicit Reasoning via Frequency-Aware Multi-Granularity Graph Flow
arXiv · 17 iun. 2026