RESEARCHInvestigateNEXT 12 MONTHS
Tokengeist: Multi-Turn Attribution Tracing in Agentic Conversations
arXiv cs.CL — Computation and Language
Factual evidence
What the source reports
Researchers introduce Tokengeist, a multi-turn attribution tracing method for agentic conversations, improving lineage tracking for LLM responses.
Inspect the evidence
- Inclusion basis
- Enterprise AI
- Publisher and source type
- arXiv cs.CL — Computation and Language · RESEARCH
- Published by source
- 20 November 2026
- Collected by OneBench
- 28 Jul 2026, 15:18 UK
Stored source excerpt
arXiv:2607.22610v1 Announce Type: cross Abstract: When a language model produces a response in a multi-turn conversation, which tokens from prior turns shaped that answer, and…
Short excerpt from the collected text, not the full source. Use the source link to read it in context.
The factual summary is a OneBench synthesis, not a quotation or independent verification. Collection time is not publication time. Open the source for its full context; related reporting can share the same underlying announcement.