self-augmenting-retrieval-for-diffusion-language-models-b1d03fa5·1 events·first seen Aliases: Self-Augmenting Retrieval for Diffusion Language Models
Researchers introduce SARDI, a training-free RAG framework for discrete diffusion language models that repurposes discarded low-confidence tokens during denoising as lookahead signals to guide retrieval before output is finalized. The method is retriever-agnostic and applicable to any reasoning-capable discrete diffusion LM. Evaluated across five multi-hop QA benchmarks, SARDI outperforms training-free diffusion and autoregressive retrieval baselines at up to 8x higher throughput.