CAMS framework extracts, clusters, selects, and rewrites atomic claims to produce multi-document summaries with fine-grained, multi-source traceability by construction.
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics , pages=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CL 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
TSM-Bench shows SOTA MGT detectors drop 10-40% in accuracy on task-specific Wikipedia edits versus generic text, with fine-tuning on task-specific data generalizing better than the reverse.
citing papers explorer
-
Faithful by Construction: Claim-Anchored Attribution for Multi-Document Summarization
CAMS framework extracts, clusters, selects, and rewrites atomic claims to produce multi-document summaries with fine-grained, multi-source traceability by construction.
-
TSM-Bench: Detecting LLM-Generated Text in Real-World Wikipedia Editing Practices
TSM-Bench shows SOTA MGT detectors drop 10-40% in accuracy on task-specific Wikipedia edits versus generic text, with fine-tuning on task-specific data generalizing better than the reverse.