Deep reasoning

I study how causality and other formal reasoning systems can improve AI reasoning and help solve complex problems that require exploration over thousands of steps. See Plan*RAG for an example on how solving complex tasks can be abstracted out as planning over a directed acyclic graph.

Improving AI reasoning
Text-based optimization for generative AI

References

  1. Arxiv
    Plan*RAG: Planning-guided Retrieval Augmented Generation
    Prakhar Verma, Sukruta Prakash Midigeshi, Gaurav Sinha, Arno Solin, Nagarajan Natarajan, and Amit Sharma
    arXiv preprint arXiv:2410.20753, 2024
    arXiv
  2. Arxiv
    Teaching Transformers Causal Reasoning through Axiomatic Training
    Aniket Vashishtha, Abhinav Kumar, Abbavaram Gowtham Reddy, Vineeth N Balasubramanian, and Amit Sharma
    arXiv preprint arXiv:2407.07612, 2024
    arXiv
  3. Arxiv
    Task Facet Learning: A Structured Approach to Prompt Optimization
    Gurusha Juneja, Nagarajan Natarajan, Hua Li, Jian Jiao, and Amit Sharma
    arXiv preprint arXiv:2406.10504, 2024
    arXiv
  4. ACL 2024
    NICE: To optimize in-context examples or not?
    Pragya Srivastava, Satvik Golechha, Amit Deshpande, and Amit Sharma
    In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2024