reward-acoustic-fidelity-frozen-judge-during-rl TEC-0180 · proposedReward acoustic fidelity with a frozen judge during RL
Augments the text-space -WER GRPO reward with a group-normalized score from a permanently frozen, non-autoregressive CTC model that measures how well each sampled transcription aligns to the audio, used at training time only.
Proposed. Filed by the drafting stage because a paper introduces or tests it. Nobody has vouched for it. The claim drafted from that paper is its first evidence, and what the paper used to validate it is in that claim’s source note. Say whether it held up for you, below.
Augments the text-space -WER GRPO reward with a group-normalized score from a permanently frozen, non-autoregressive CTC model that measures how well each sampled transcription aligns to the audio, used at training time only. Filed by the drafting stage from Grounded in Sound: Reinforcement Learning with a Frozen Acoustic Judge to Curb ASR Insertion Hallucinations (2609.14455) with status proposed: the paper introduces or tests it, and the claim drafted from the paper is its first evidence. The paper validated it by: Insertion counts and WER on 33,282 utterance-condition instances over six difficulty tiers against a schedule-matched mu=0 WER-GRPO baseline sharing identical data and schedule; meeting-level clustered bootstrap CIs, three seeds, leave-one-meeting-out, plus counterfactual probes on unintelligible audio, a 32-best CTC rescoring comparison, and insertion-AURC under policy-only confidence.. Nobody has vouched for it; add support or contest it on this page.
Does it work?
nothing measured0 supporting · 0 contesting sources
Efficacy claims — what this technique actually moves, under which conditions, and whether that has been contested.
No efficacy claim filed yet. The technique is catalogued; whether it moves the capability, and when, is a separate assertion that needs its own sources.
No search recorded either, so this says nothing about the literature — only that nobody has looked here yet.
Code
No repository linked yet. Contribute one.
Sources
Whether a technique works is a claim. What you file here becomes a claim about this technique under the capability it addresses, with your evidence as its source, and moves the standing above the same way a paper would.