Spaces:
Running
Running
Organization card
Browse files
README.md
CHANGED
|
@@ -1,13 +1,36 @@
|
|
| 1 |
---
|
| 2 |
title: README
|
| 3 |
-
emoji:
|
| 4 |
colorFrom: gray
|
| 5 |
-
colorTo:
|
| 6 |
-
sdk:
|
| 7 |
-
sdk_version: 6.30.0
|
| 8 |
-
python_version: '3.13'
|
| 9 |
-
app_file: app.py
|
| 10 |
pinned: false
|
| 11 |
---
|
| 12 |
|
| 13 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
title: README
|
| 3 |
+
emoji: 馃攳
|
| 4 |
colorFrom: gray
|
| 5 |
+
colorTo: gray
|
| 6 |
+
sdk: static
|
|
|
|
|
|
|
|
|
|
| 7 |
pinned: false
|
| 8 |
---
|
| 9 |
|
| 10 |
+
# TraceLite AI
|
| 11 |
+
|
| 12 |
+
An academic-driven open research team working on the building blocks of AI agent evaluation and post-training: **forensics, attribution, environments, rewards, and data generation**. Each component can be used on its own for model and harness selection, training, and pre-release acceptance testing, and together they form a framework for continual self-improvement.
|
| 13 |
+
|
| 14 |
+
[Website](https://tracelite.ai) 路 [GitHub](https://github.com/TraceLite-AI) 路 [Research timeline](https://tracelite.ai/timeline.html) 路 contact@tracelite.ai
|
| 15 |
+
|
| 16 |
+
## Trial by Trace 路 TraceTrial-Bench
|
| 17 |
+
|
| 18 |
+
When an AI agent fails, was it the model or the harness around it? TraceTrial-Bench scores execution traces in two layers, task outcome and harness reliability, across 95 mechanism-oriented tasks, 4 harnesses (Claude Code, Codex, OpenClaw, DeepSeek Harness) and 9 models, 3,420 runs in total.
|
| 19 |
+
|
| 20 |
+
- 馃攳 [Interactive report and leaderboard (Space)](https://huggingface.co/spaces/TraceLite-AI/Trial-by-Trace)
|
| 21 |
+
- 馃捇 [Code and tasks](https://github.com/TraceLite-AI/TraceTrial-Bench)
|
| 22 |
+
- 馃搫 [Technical report](https://tracelite.ai/trial-by-trace.html)
|
| 23 |
+
|
| 24 |
+
## Papers (2026)
|
| 25 |
+
|
| 26 |
+
| Paper | Venue |
|
| 27 |
+
|---|---|
|
| 28 |
+
| [MM-VeriRubric: Verifiable Evidence-Grounded Rubrics for Multimodal Reward Models](https://tracelite.ai/papers.html) | EMNLP 2026 |
|
| 29 |
+
| [Disentangled Privileged States for Simulator-Assisted Multi-Turn Dialogue Policy Training](https://tracelite.ai/papers.html) | EMNLP 2026 Findings |
|
| 30 |
+
| [Absorbing Gradient Conflicts: Modeling Semantic Variance via Kent Distributions for Cross-Modal Hashing](https://arxiv.org/abs/2608.24010) | IJCAI 2026 |
|
| 31 |
+
| [SimRPD: Optimizing Recruitment Proactive Dialogue Agents through Simulator-Based Data Evaluation and Selection](https://aclanthology.org/2026.acl-industry.95/) | ACL 2026 路 Oral |
|
| 32 |
+
| [Look Less, Reason More: Rollout-Guided Adaptive Pixel-Space Reasoning](https://aclanthology.org/2026.acl-long.225/) | ACL 2026 |
|
| 33 |
+
| [Select Less, Reason More: Prioritizing Evidence Purity for Video Reasoning](https://openaccess.thecvf.com/content/CVPR2026/html/Li_Select_Less_Reason_More_Prioritizing_Evidence_Purity_for_Video_Reasoning_CVPR_2026_paper.html) | CVPR 2026 |
|
| 34 |
+
| [CausalStep: A Benchmark for Explicit Stepwise Causal Reasoning in Videos](https://arxiv.org/abs/2507.16878) | AAAI 2026 路 Oral |
|
| 35 |
+
|
| 36 |
+
Full list: [tracelite.ai/papers.html](https://tracelite.ai/papers.html)
|