arXiv AI September 18, 2026 By Donghan Bian (ENC, LRE), Marie Puren (LRE, ENC), Florian Cafiero (LRE, ENC)

TRACE: Accountable Agentic Retrieval for Source Discovery in Digital Archives

TRACE is a training‑free, agentic retrieval framework that enables accountable source discovery in historical archives, addressing challenges such as OCR degradation and genre heterogeneity. Developed for the DECIDON project on French Third Republic political discourse, it is deployed internally for 24 researchers across six institutions. On the HistoriQA‑ThirdRepublic benchmark, TRACE achieves R@10 of 0.856 and MRR of 0.653, outperforming sparse, dense, graph‑based, and other agentic RAG baselines, especially on multi‑hop and cross‑corpus questions, while costing only about $0.02 per question.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv AI

Jul 1

HistoriQA-ThirdRepublic: Multi-Hop Question Answering Corpus for Historical Research, Parliamentary Debates from the French Third Republic (1870-1940)

arXiv:2606. 31325v1 Announce Type: new Abstract: We present HistoriQA-ThirdRepublic: a French-language dataset of multi-hop historical questions derived from parliamentary debates and newspapers of the French Third Republic.

By Aur\'elien Pellet (LRE), Julien Perez (EPITA, LRE), Marie Puren (LRE, CJM)

llms rag nlp benchmarks safety

arXiv Computation and Language

Sep 11

INDRA: A New AI Tool for Exploring Tobacco, Fossil Fuel, and Chemical Industry Archives

INDRA is a research platform that integrates multiple archival collections—such as UCSF’s Industry Documents Library, Columbia and CUNY’s ToxicDocs, and Stanford’s SRITA—into a single, LLM‑readable corpus. It employs three safeguards: a closed evidentiary sandbox, real‑time provenance tagging, and a deterministic system‑level protocol to ensure that model outputs are clearly distinguished from archival evidence and from the model’s own inferences. The platform enables large‑language‑model‑powered investigations across these archives while keeping the conditions of knowledge production transparent and auditable.

By Daniel Akselrad, Robert N. Proctor

llms rag safety

Hugging Face Trending Papers

Jun 23

AGORA: An Archive-Grounded Benchmark for Agentic Workplace Document Reasoning

Large language models are increasingly deployed as agents that reason over documents rather than answer from parametric knowledge. We study archive-grounded reasoning: locating sparse evidence across a large, messy collection of workplace files, reconciling inconsistent terminology, units, and time conventions, and computing an answer.

llms agents benchmarks

arXiv Computation and Language

Sep 1

Extending AI for Research to the Humanities: A Multi-Agent Framework for Evidence-Grounded Scholarship

arXiv:2605.30947v4 Announce Type: replace Abstract: LLM-based research agents have advanced rapidly in science and engineering, where research is organized around executable experiments, code, and qu...

By Yating Pan, Jiajun Zhang, Jun Wang, Qi Su

llms rag agents benchmarks

arXiv AI

Jul 28

VecTree-RAG: An Agentic Retrieval-Augmented Generation Framework Combining Vector and Tree Retrieval for Efficiency and Accuracy

arXiv:2607. 23006v1 Announce Type: cross Abstract: Scientific question answering requires a retrieval system to solve two distinct problems: identifying which papers are relevant and locating the supporting evidence within those papers.

By Xinyan Zhong, Yuwei Shi, Yuqi Wei, Chen Shen, Tianhang Zhou, Zhenghao Wu

llms rag agents nlp benchmarks

arXiv Machine Learning

Jul 31

EMBL AI Librarian: Life-Sciences Knowledge Layer for AI Agents

arXiv:2607. 28229v1 Announce Type: cross Abstract: The web is increasingly accessed by AI agents rather than humans.

By Luigi Sigillo, Matteo Silvestri, Francesco Tabaro, Rajat Bhatnagar, Syed Irtaza Mubashar, Matt Jeffryes, Daljit Nijjer, Vittorio Perera, Ola Spjuth, Julio Saez-Rodriguez, Melissa Harrison, Fabio Petroni

llms agents nlp robotics benchmarks