Trace Integrity for LLM Data Agents: A Vision for Auditable Structured Reasoning in Real-World Systems
Answer accuracy is an insufficient reliability signal for LLM data agents. In structured-data tasks, a benchmark-correct answer can be produced by an invalid trace. This paper introduces Trace Integrity, a deployment reliability criterion for evaluating whether the computation recorded behind an answer is explicit, executable, schema-valid, operator-faithful, replayable, answer-consistent, and auditable. We identify the Structure Gap as the deployment failure mode that makes Trace Integrity necessary: natural-language reasoning and free-form rationales do not reliably specify the operator-leve
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- FuzzySimilar title/name (fuzzy) · 59%VioletVision-3B →
“Fuzzy title match (0.73): “Trace Integrity for LLM Data Agents: A Vision for Auditable ” ≈ “VioletVision-3B””
- LinkedLinked via arxiv author · 85%Srimonti Dutta →
“Trace Integrity for LLM Data Agents: A Vision for Auditable Structured Reasoning in Real-World Systems”
- LinkedLinked via arxiv author · 85%Akshata Kishore Moharir →
“Trace Integrity for LLM Data Agents: A Vision for Auditable Structured Reasoning in Real-World Systems”
- FuzzySimilar title/name (fuzzy) · 87%NirDiamant/GenAI_Agents →
“Fuzzy title match (0.94): “Trace Integrity for LLM Data Agents: A Vision for Auditable ” ≈ “NirDiamant/GenAI_Agents””
- FuzzySimilar title/name (fuzzy) · 84%Unity-Technologies/ml-agents →
“Fuzzy title match (0.92): “Trace Integrity for LLM Data Agents: A Vision for Auditable ” ≈ “Unity-Technologies/ml-agents””
- FuzzySimilar title/name (fuzzy) · 84%pytorch/vision →
“Fuzzy title match (0.92): “Trace Integrity for LLM Data Agents: A Vision for Auditable ” ≈ “pytorch/vision””
- FuzzySimilar title/name (fuzzy) · 59%datawhalechina/hello-agents →
“Fuzzy title match (0.73): “Trace Integrity for LLM Data Agents: A Vision for Auditable ” ≈ “datawhalechina/hello-agents””
- FuzzySimilar title/name (fuzzy) · 59%Eigenwise/atomic-agents →
“Fuzzy title match (0.73): “Trace Integrity for LLM Data Agents: A Vision for Auditable ” ≈ “Eigenwise/atomic-agents””