{"slug":"agentscope-ai-02-rl-reward","source_name":"agentscope-ai/02-rl-reward","name":"Agentscope AI/02 Rl Reward","description":"Build RL reward signals using the OpenJudge framework. Covers choosing between pointwise and pairwise reward strategies based on RL algorithm, task type, and cost; aggregating multi-dimensional pointwise scores into a scalar reward; pairwise tournament reward for GRPO on subjective tasks (net win rate across group rollouts); generating preference pairs for DPO/RLAIF; and normalizing scores for training stability. Use when building reward models, scoring rollouts for GRPO/REINFORCE, generating pr","version":1,"lift":{"pass_rate_delta_pts":26.92,"pass_rate_pct":92.3,"total_cases":26,"passed_cases":24,"tokens_delta_pct":35.4,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-10-11T00:39:47.720739+00:00"},"skill_score":0.9231,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":26.92,"with_pass_pct":92.3,"without_pass_pct":65.4,"tokens_delta_pct":35.4,"turns_delta_pct":0,"total_cases":26,"cases_aggregated":26,"verdict":"mixed","never_hurt":false,"completed_at":"2026-10-11T00:39:47.720739+00:00","run_id":"f05446e3-442a-48d4-9553-8657eb78b496","latest_attempt":{"id":"f05446e3-442a-48d4-9553-8657eb78b496","runner_model":"gemini-3.6-flash","overall_verdict":"mixed","completed_at":"2026-10-11T00:39:47.720739+00:00","gate":null},"version_number":1,"is_latest_version":true,"gate":null}],"trust":{"skill_safety":"passed","safety_status":"clean","intent_verdict":"safe","content_status":"clean","indexable":true},"license":"Apache-2.0","install_count":0,"total_uses":0,"total_users":0,"manifest_hash":"b38134de3c6959b3fe206faa787fa58e75dd908c2dad6a4661ebe40006f2d17a","install_name":"agentscope-ai-02-rl-reward","version_id":"5473b901-8ea8-46a8-91d3-785912530a45","body_sha256":"b38134de3c6959b3fe206faa787fa58e75dd908c2dad6a4661ebe40006f2d17a","render_hash":"2fd72bb81f55542242ed81ac918221105d0f1e331d793d972b948a5ab9d06d84","raw_url":"https://app.decimal.ai/s/agentscope-ai-02-rl-reward/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/agentscope-ai-02-rl-reward"}