Metraipolis › archive › dc_6edd063cfe05
SearchAtlas: Analyzing Agentic Search Strategies via Evidential Query Graphs — arXiv:2609.10901
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.10901
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T05:40:40.387Z
- SHA-256 of the archived text
8254ca26d6976a8081aabea0da7d13ec70c9a0426d045710cbd411608a4395b2- Stable URL of this page
https://metraipolis.com/documento/dc_6edd063cfe05
How to cite
SearchAtlas: Analyzing Agentic Search Strategies via Evidential Query Graphs — arXiv:2609.10901 — arXiv (en), https://arxiv.org/abs/2609.10901 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6edd063cfe05 (Metraipolis archive dc_6edd063cfe05, sha256 8254ca26d6976a8081aabea0da7d13ec70c9a0426d045710cbd411608a4395b2)
BibTeX
@misc{dc6edd063cfe05,
title = {SearchAtlas: Analyzing Agentic Search Strategies via Evidential Query Graphs — arXiv:2609.10901},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.10901},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6edd063cfe05; sha256 8254ca26d6976a8081aabea0da7d13ec70c9a0426d045710cbd411608a4395b2},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.10901. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
LLM search agents are often evaluated on final-answer accuracy, overlooking the process. Analyzing a search strategy requires understanding how credible evidence is retrieved to address question constraints. This valuable information is buried in raw search trajectories that are long and difficult to parse. We introduce SearchAtlas, a framework that converts search trajectories into structured graphs whose edges represent how evidence is propagated across the reasoning trace, from the query that retrieves it to the final answer. Our automated parsing pipeline achieves a mean edge F1 of 86.0% against human-annotated graphs and remains consistent across repeated runs. We analyze five search agents on three benchmarks, revealing systematic differences in search scale and evidence aggregation. SearchAtlas exposes fragmented answer support, question constraints that do not reach the answer, and unverified parametric knowledge entering the response. These process failures are strongly associated with incorrect answers, even more so than an LLM judge given either the raw trajectory or the ordered query list, suggesting that the constructed graphs provide useful interpretability. Moreover, an audit of cases in which process-diagnostic scores disagree with final-answer correctness shows that they capture information not reducible to answer accuracy. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-11 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.10901 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-16T05:23:56.984Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-11?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).