AI Paper Atlas

Research map / Large language models / Agents, code and math

ADEPT: RL-Aligned Agentic Decoding of Emotion via Evidence Probing Tools — From Consensus Learning to Ambiguity-Driven Emotion Reasoning

Esther Sun, Bo-Hao Su, Abinay Reddy Naini, Shinji Watanabe, Carlos Busso

ICML 2026 · Large language models › Agents, code and math · approach: Long-horizon LLM agents · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add the ADEPT tool protocol to Qwen-3-Omni (without GRPO) (ablation)Qwen-3-Omni · MSP-Podcast Test1Set Recall: 49.23 to 57.48Improved
add GRPO alignment to ADEPT (ablation)ADEPT · MSP-Podcast Test1Primary Macro-F1: 35.71 -> 42.24Improved
add GRPO alignment to ADEPT (ablation)ADEPT · MSP-Podcast Test1Jaccard: 44.50 -> 47.51Improved
add prior tools (ablation)ADEPT · MSP-Podcast Test1Set Recall: 58.17 -> 61.92Improved
add semantic probing tools (ablation)ADEPT · MSP-Podcast Test1Primary Macro-F1: 33.86 -> 42.24Improved
add acoustic probing tools (ablation)ADEPT · MSP-Podcast Test1Jaccard: 41.28 -> 47.51Improved
add replay tool (ablation)ADEPT · MSP-Podcast Test1Primary Macro-F1: 40.44 -> 42.24Improved
add Evidence Trust Gate (ablation)ADEPT · MSP-Podcast Test1Jaccard: 44.96 -> 47.51Improved
ADEPT + GRPO versus fine-tuned WavLMADEPT · MSP-Podcast Test1Primary Macro-F1: 0.3640 -> 0.4224Improved
zero-shot ADEPT + GRPO versus supervised BLSP-EmoADEPT · IEMOCAP (zero-shot, mapped 7-way)Set Recall: 52.30 -> 54.80Improved
zero-shot ADEPT + GRPO versus supervised BLSP-EmoADEPT · IEMOCAP (zero-shot, mapped 7-way)Accuracy: 76.13 -> 54.83Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy