Dataset Viewer
Auto-converted to Parquet Duplicate
type
stringclasses
2 values
harness
stringclasses
1 value
id
stringclasses
1 value
name
stringclasses
1 value
message
dict
session
trackio
agent_run
Agent repro run 2026-08-02 00:41 UTC
null
message
null
null
null
{ "role": "system", "content": "Event", "timestamp": 1785631290065 }
message
null
null
null
{ "role": "assistant", "content": "[Claim 0 — DPIMF framework for differentially private implicit MF]\n---\nclaim: 0\npaper_id: Qo7ENqEdq9\njob_id: 6a5eebcd13e6ef894d5494fa\n---\n\n# Claim 0 — DPIMF framework for differentially private implicit MF\n\n> \"Proposes DPIMF framework for differentially private implicit ...
message
null
null
null
{ "role": "assistant", "content": "[Claim 1 — Loss redesign + importance sampling reduces noise scale for privacy]\n---\nclaim: 1\npaper_id: Qo7ENqEdq9\njob_id: 6a5eebd013e6ef894d5494fc\n---\n\n# Claim 1 — Loss redesign + importance sampling reduces noise scale for privacy\n\n> \"Redesigns loss function and adopts ...
message
null
null
null
{ "role": "assistant", "content": "[Claim 2 — Better privacy-utility trade-off on three benchmark datasets]\n---\nclaim: 2\npaper_id: Qo7ENqEdq9\njob_id: 6a5eebd4d09dc1f57c6be7d9\n---\n\n# Claim 2 — Better privacy-utility trade-off on three benchmark datasets\n\n> \"Achieves better privacy-utility trade-off than st...
message
null
null
null
{ "role": "assistant", "content": "[DPIMF — Differentially Private Implicit Matrix Factorization (paper Qo7ENqEdq9)]\n---\ntitle: \"Reproducing Qo7ENqEdq9 — DPIMF (Differentially Private Implicit MF)\"\npaper_id: Qo7ENqEdq9\n---\n\n# DPIMF — Differentially Private Implicit Matrix Factorization (paper Qo7ENqEdq9)\n\...
message
null
null
null
{ "role": "system", "content": "Event", "timestamp": 1785631290065 }

Agent traces

Agent sessions published from a Trackio Logbook.

Downloads last month
31