sumitrail commited on
Commit
b4eee60
·
verified ·
1 Parent(s): bbacf2d

docs: add RAIL Guard paper (arXiv:2607.16215) to research section

Browse files
Files changed (1) hide show
  1. README.md +2 -0
README.md CHANGED
@@ -34,6 +34,8 @@ https://mcp.responsibleailabs.ai/mcp
34
 
35
  ## Research
36
 
 
 
37
  **RAIL in the Wild** (arXiv:2505.00204): Operationalizing responsible AI evaluation using Anthropic's Values in the Wild dataset (308,000+ conversations). Maps AI-expressed values to RAIL's 8 dimensions with quantitative scoring. [Read the paper](https://arxiv.org/abs/2505.00204).
38
 
39
  ## Links
 
34
 
35
  ## Research
36
 
37
+ **RAIL Guard: Closing the Evaluation-to-Remediation Gap in Responsible AI for LLM Agents** (arXiv:2607.16215): A closed-loop responsible AI pipeline that evaluates LLM outputs across 8 measurable dimensions and iteratively remediates failing outputs through an evaluate-rewrite-reevaluate loop, evaluated on four frontier LLMs using the [RAIL Guard Benchmark](https://huggingface.co/datasets/responsible-ai-labs/rail-guard-benchmark). [Read the paper](https://arxiv.org/abs/2607.16215).
38
+
39
  **RAIL in the Wild** (arXiv:2505.00204): Operationalizing responsible AI evaluation using Anthropic's Values in the Wild dataset (308,000+ conversations). Maps AI-expressed values to RAIL's 8 dimensions with quantitative scoring. [Read the paper](https://arxiv.org/abs/2505.00204).
40
 
41
  ## Links