9 papers
OpenSIR: Open-Ended Self-Improving Reasoner
Wai-Chung Kwan, Joshua Ong Jun Leang, Pavlos Vougiouklis +3
Recent advances in large language model (LLM) reasoning through reinforcement learning rely on annotated datasets for verifiable rewards, which may limit models' ability to surpass…
From An LLM Swarm To A PDDL-Empowered HIVE: Planning Self-Executed Instructions In A Multi-Modal Jungle
Kaustubh Vyas, Damien Graux, Yijun Yang +8
In response to the call for agent-based solutions that leverage the ever-increasing capabilities of the deep models' ecosystem, we introduce Hive -- a comprehensive solution for kn…
Millions of -s: Extending GraphRAG to Millions of Documents
Zhili Shen, Chenxin Diao, Pascual Merita +2
Recent studies have explored graph-based approaches to retrieval-augmented generation, leveraging structured or semi-structured information -- such as entities and their relations…
GeAR: Graph-enhanced Agent for Retrieval-augmented Generation
Zhili Shen, Chenxin Diao, Pavlos Vougiouklis +12
Retrieval-augmented Generation (RAG) relies on effective retrieval capabilities, yet traditional sparse and dense retrievers inherently struggle with multi-hop retrieval scenarios.…
An Extensive Evaluation of PDDL Capabilities in off-the-shelf LLMs
Kaustubh Vyas, Damien Graux, Sébastien Montella +5
In recent advancements, large language models (LLMs) have exhibited proficiency in code generation and chain-of-thought reasoning, laying the groundwork for tackling automatic form…
How Reliable are LLMs as Knowledge Bases? Re-thinking Facutality and Consistency
Danna Zheng, Mirella Lapata, Jeff Z. Pan
Large Language Models (LLMs) are increasingly explored as knowledge bases (KBs), yet current evaluation methods focus too narrowly on knowledge retention, overlooking other crucial…