Back to News

Continuous Diffusion Language Models Introduced for Text Generation

134 points · 48 comments#diffusion#language-model#text-generation#research

A Hacker News post highlights Continuous Diffusion Language Models (CDLM), a research approach that applies continuous diffusion to language modeling. The method is presented as an alternative to autoregressive models for text generation.

Coverage timeline

  1. Hacker Newspeter_d_sherman
    Continuous Diffusion Language Models (CDLM's)HN

Related stories

  • Paper Proposes Direct Semantic Communication Between LLMs via Cache-to-Cache

    A 2025 paper, discussed on Hacker News, proposes a method for direct semantic communication between large language models by transferring cache states, bypassing traditional text-based exchanges. The approach aims to enable more efficient and nuanced information sharing between models.

    1 reports · 1 sources107 points · 20 comments

  • WebLLM: High-Performance In-Browser LLM Inference Engine Showcased

    WebLLM, a high-performance in-browser LLM inference engine, was showcased on Hacker News on September 2, 2026. The engine enables large language model inference directly within web browsers, potentially reducing server-side costs and enhancing privacy. This demonstration highlights the growing feasibility of running sophisticated AI models client-side.

    1 reports · 1 sources145 points · 24 comments

  • Hacker News Thread Discusses Extensible Software Design for LLM Applications

    A Hacker News discussion on August 19, 2026, examined how to design extensible software systems in the context of LLM-based applications. The thread focused on architectural patterns and engineering practices for building adaptable systems that integrate large language models.

    1 reports · 1 sources171 points · 88 comments

  • Researchers Steal Encrypted Reasoning Traces from Claude, GPT, Gemini APIs

    A new paper demonstrates a method to extract hidden chain-of-thought reasoning from proprietary LLM APIs by replaying encrypted reasoning blocks into weaker sibling models from the same provider, which then output the traces in plaintext. The researchers verified their extracted token counts match billed API thinking tokens 1:1 for most prompts, suggesting near-complete coverage of hidden reasoning. The vulnerability affects Anthropic, OpenAI, and Google models, with cross-model compatibility allowing traces to be replayed across sessions, users, and models.

    5 reports · 4 sources694 points · 306 comments

News updates every hour. Subscribe in your RSS reader to get new stories automatically.

Subscribe