Research

Articles filed under Research.

Charles Gao presents Hill Research's ESDA paper at the 32nd ACM SIGKDD Conference at ICC Jeju in Jeju, Korea
Research

Hill Research Presents ESDA at KDD 2026 in Jeju, Korea

Hill Research presented "From Tool Traces to Strategy Banks" at the 32nd ACM SIGKDD Conference in Jeju, Korea, introducing ESDA — a system that mines agent debugging traces into a reusable strategy bank and reaches 84.7% success on RepoBench with 4.0x fewer evaluator calls.

Read →
Kai Zheng of Hill Research beside the poster at board G17, ACL 2026, San Diego
Research

Hill Research at ACL 2026: Contract-Checked Editing for Verifier-Guided LLM Reasoning

Hill Research's Kai Zheng presented "From Trajectories to Graphs: Contract-Checked Editing for Verifier-Guided LLM Reasoning" at the 64th Annual Meeting of the Association for Computational Linguistics in San Diego. The long paper lifts verifier-runnable recombination from 41.2% to 92.8% while using 42% fewer verifier calls.

Read →
Hill Research publishes five papers at ACL, SIGMETRICS, MLSys, and JAMIA Open in April 2026
Research

Five Papers Accepted at ACL, SIGMETRICS, MLSys, and JAMIA in One Week

Hill Research's CTO Dr. Jack Li had four papers accepted at top computer science venues — two at ACL 2026, one at SIGMETRICS 2026, and one at MLSys 2026 — plus a peer-reviewed publication in JAMIA Open describing ClinicalMind, the knowledge graph layer underneath TriClick.

Read →
Building Trust in Medical AI - Inside Med-ICE Podcast Episode
Research

Podcast: Building Trust in Medical AI — Inside Med-ICE

In this episode of AI on the Hill in the Spotlight, Zhiyuan Chen discusses Med-ICE, a multi-agent consensus framework designed to improve the trustworthiness and reliability of medical AI by aligning reasoning across multiple LLM agents.

Read →
Hill Research at AAAI 2026 SPARTA Workshop in Singapore
Research

Hill Research Presents at AAAI 2026 SPARTA Workshop in Singapore

Hill Research's AI team presented three research projects at the SPARTA workshop during AAAI 2026, covering cross-species transfer learning, dynamic gene expression modeling, and reducing LLM hallucinations in clinical AI.

Read →
AI-Powered Framework for Cross-Species Single-Cell Analysis
Research

Hill Research New Preprint: AI-Powered Framework for Cross-Species Single-Cell Insight

Hill Research announces a new preprint on BioRxiv introducing CSLAN, a transfer learning framework that uses mouse scRNA-seq data to identify human trauma-related immune cells with 96.67% accuracy.

Read →