# DelTA: Discriminative Token Credit Assignment for Reinforcement Learning from Verifiable Rewards

Page: https://stenobird.com/podcast/daily-paper-cast-7079649/delta-discriminative-token-credit-assignment-for-reinforcement-learning-from-verifiable-rewards
Text version: https://stenobird.com/podcast/daily-paper-cast-7079649/delta-discriminative-token-credit-assignment-for-reinforcement-learning-from-verifiable-rewards.md
Podcast: [Daily Paper Cast](https://stenobird.com/podcast/daily-paper-cast-7079649)
Published: 2026-05-23T04:29:01+00:00
Episode link: https://share.transistor.fm/s/428a7c58
Audio file: https://media.transistor.fm/428a7c58/929d124c.mp3
Processing state: not_requested
JSON: https://stenobird.com/v1/public/podcasts/daily-paper-cast-7079649/episodes/delta-discriminative-token-credit-assignment-for-reinforcement-learning-from-verifiable-rewards
Duration seconds: 1286

## Resource

🤗 Upvotes: 124 | cs.LG, cs.CL Authors: Kaiyi Zhang, Wei Wu, Yankai Lin Title: DelTA: Discriminative Token Credit Assignment for Reinforcement Learning from Verifiable Rewards Arxiv: http://arxiv.org/abs/2605.21467v1 Abstract: Reinforcement learning from verifiable rewards (RLVR) has emerged as a central technique for improving the reasoning capabilities of large language models. Despite its effectiveness, how response-level rewards translate into token-level probability changes remains poorly understood. We introduce a discriminator view of RLVR updates, showing that the policy-gradient update direction implicitly acts as a linear discriminator over token-gradient vectors and thereby determines which token probabilities are increased or decreased during learning. Under standard sequence-level RLVR, this discriminator is constructed from positive- and negative-side centroids formed by advantage-weighted averaging of token-gradient vectors. However, such centroid construction can be dominated by shared high-frequency patterns, such as formatting tokens, diluting sparse yet discriminative directions that better distinguish high-reward responses from low-reward ones. To address this limitation, we propose $\textbf{DelTA}$, a discriminative token credit assignment method that estimates token coefficients to amplify side-specific token-gradient directions and downweight shared or weakly discriminative ones. These coefficients reweight a self-normalized RLVR surrogate, making the effective side-wise centroids more contrastive and thereby reshaping the RLVR update direction. On seven mathematical benchmarks, DelTA outperforms the strongest same-scale baselines by 3.26 and 2.62 average points on Qwen3-8B-Base and Qwen3-14B-Base, respectively. Additional results on code generati…

## Actions

- request_transcript: `POST https://stenobird.com/v1/public/podcasts/daily-paper-cast-7079649/episodes/delta-discriminative-token-credit-assignment-for-reinforcement-learning-from-verifiable-rewards/transcription-requests` — Idempotently request low-priority transcript generation for this episode.
- read_markdown: `GET https://stenobird.com/podcast/daily-paper-cast-7079649/delta-discriminative-token-credit-assignment-for-reinforcement-learning-from-verifiable-rewards.md` — Read the agent-friendly Markdown representation of this episode resource.

A page view does not enqueue transcription. Agents should invoke `request_transcript` explicitly when they need this episode processed.

## Transcript

Full transcripts are not published on public pages unless there is a clear rights basis.