AI Paper Atlas

Research map / Large language models / Language, safety and interpretability

SageLM: A Multi-aspect and Explainable Large Language Model for Speech Judgement

Yuan Ge, Junxiang Zhang, Xiaoqian Liu, Bei Li, Xiangnan Ma, Chenglong Wang, Kaiyang Ye, Yangfan Du, Linfeng Zhang, Yuxin Huang, Tong Xiao, Zhengtao Yu and 1 more

AAAI 2026 · Large language models › Language, safety and interpretability · approach: Speech and audio language models

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
fine-tune with rationale-augmented SFT instead of label-only SFT or GRPO (ablation)speech judge (Qwen2.5-omni) · SpeechFeedback preliminarycompare accuracy / agreementImproved
use GRPO rule-based RL instead of label-only SFT (ablation)speech judge · SpeechFeedback preliminary, scoring accuracyscoring accuracyImproved
use GRPO instead of label-only SFT (ablation)speech judge · SpeechFeedback preliminary, pairwise comparecompare accuracyNo clear effect
train with GRPO without supervising rationales (ablation)speech judge · manual analysis of 100 sampled casesrationale-score consistency: explanation contradicted final score in 39% of casesHurt
add stage 1 semantic training only (GRPO or SFT) (ablation)speech judge · acoustic evaluationacoustic agreementHurt
add stage 2 acoustic preference data after stage 1 (ablation)SageLM · acoustic and semantic evaluationagreementImproved
end-to-end SageLM instead of Whisper + GPT-4o cascadespeech judge · AlpacaEval real S2S outputsagreement: 69.30% -> 87.97%Improved
two-stage SFT on Qwen2.5-omni-7B vs direct promptingQwen2.5-omni-7B · SpeechFeedback semantic test, Helpfulnessagreement (%): 56.50 -> 83.61Improved
use task-specific PandaLM-7B vs general Qwen2.5-32B in cascade (ablation)Whisper cascade judge · SpeechFeedback semantic testagreement (%): 67.58 vs 67.40 (Helpfulness)No clear effect
scale base model from 3B to 7B for SageLM (ablation)SageLM · SpeechFeedback acoustic test, Mixedagreement (%): 94.44 -> 91.36 (Mixed); 97.47 -> 98.65 (Implicit)Mixed
increase combined response pair length beyond 60 s (ablation)SageLM · AlpacaEval long responsesagreement with humansHurt

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