SKG-Eval: Stateful Evaluation of Multi-Turn Dialogue via Incremental Semantic Knowledge Graphs

Avijit Shil Suman Samui, [1ex] Maulana Abul Kalam Azad University of Technology, West Bengal, India · 2026

Read SKG-Eval: Stateful Evaluation of Multi-Turn Dialogue via Incremental Semantic Knowledge Graphs by Avijit Shil Suman Samui, [1ex] Maulana Abul Kalam Azad University of Technology, West Bengal, India (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-16650.txt.

Description

Shil and Samui propose SKG-Eval, a framework for judging multi-turn dialogue by tracking a growing knowledge graph of entities, relations, and commitments. It scores three things: fit to the current prompt, fit to past context, and logical coherence, the last via a geometric contradiction engine that needs no language-model judge. The three signals fuse into a session score that stays fair across lengths, with explicit contradiction certificates. Across benchmarks it matches human judgments better and catches long-range inconsistencies more reliably than flat evaluators.

Catalog record

CREATORAvijit Shil Suman Samui, [1ex] Maulana Abul Kalam Azad University of Technology, West Bengal, India
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE105917
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2605.16650

Browse the full library · About the Tower of Babel