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 2026-08-03 03:38 UTC
null
message
null
null
null
{ "role": "system", "content": "Event", "timestamp": 1785728303892 }
message
null
null
null
{ "role": "assistant", "content": "[Claim 1 — Contextual latent encoding grounds multiple referring regions]\n# Claim 1 — Contextual latent encoding grounds multiple referring regions\n\n**HF Job:** `Edd16/6a5f35ecd09dc1f57c6bef55` — COMPLETED.\n\nReal output from job logs:\n```json\n{\"contextual\": {\"recall\": {...
message
null
null
null
{ "role": "assistant", "content": "[Claim 2 — Superior multi-region referring performance]\n# Claim 2 — Superior multi-region referring performance\n\n**HF Job:** `Edd16/6a5f35ecd09dc1f57c6bef55` — COMPLETED.\n\nToy proxy aggregate performance:\n- contextual recall mean: 0.9999463538956027\n- independent recall mea...
message
null
null
null
{ "role": "assistant", "content": "[Claim 3 — Contextual latent representations handle clutter better]\n# Claim 3 — Contextual latent representations handle clutter better\n\n**HF Job:** `Edd16/6a5f35ecd09dc1f57c6bef55` — COMPLETED.\n\nToy proxy background-clutter metrics:\n- contextual bg_confusion mean: 0.0371444...
message
null
null
null
{ "role": "assistant", "content": "[Overview — Reproducing wLE7pnypx2 (ICML 2026)]\n# Overview — Reproducing wLE7pnypx2 (ICML 2026)\n\nThis logbook evaluates a numpy proxy for multi-region referring with contextual latent representations.\n\n**HF Job:** `Edd16/6a5f35ecd09dc1f57c6bef55` (cpu-basic, `pytorch/pytorch:...
message
null
null
null
{ "role": "system", "content": "Event", "timestamp": 1785728303892 }

Agent traces

Agent sessions published from a Trackio Logbook.

Downloads last month
45