• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
AI
Report

A telescopic language model aims to be valid at every depth

A post shares an excerpt describing a nested-capacity Transformer trained with a randomly truncated capacity prefix alongside a full-capacity pass.

Tanishq Mathew Abraham, Ph.D.TM
Danielle Fong 🔆DF
2 Sources, 11d ago, first seen 11d ago

TLDR

The shared excerpt says training uses one randomly truncated capacity prefix per step, trained against the full next-token target, alongside a full-capacity pass. It claims the resulting model is a valid language model at every depth.

Combined views

5.9K

2 Sources, first seen 11d ago

89 likes4 comments69 saves13 reposts

Combined views

5.9K

2 Sources, first seen 11d ago

89 likes4 comments69 saves13 reposts

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Featured Source

Sentiment

Positive——Negative

Summary

Not enough discussion yet.

No sentiment analysis available yet.

Related

How likely is AI to completely wipe us out?

A user puts their “P(doom)” at 1e-5 and calls themself an optimist.

Open models and scientists' roles in AI for science

A panelist says a September 2026 session at San Francisco's Open-Source AI Summit discussed how open models can accelerate scientific discovery and how scientists should use AI.

Tests suggest cheap verifiers can work well in LLM post-training

A post quotes a paper reporting that, in Qwen3 tests on HealthBench and PRBench, higher agreement with frontier LLM reference judges did not consistently identify the best training verifier.

2 Sources

Tanishq Mathew Abraham, Ph.D.@iScienceLuvrTelescopic Language Models "We train a Telescopic Language Model (TLM) to be that continuum: a nested-capacity Transformer supervised by stochastic prefix supervision with a full anchor. At every step, one randomly truncated prefix of the capacity axis is trained against the full next-token target, alongside one full-capacity pass, so the trained artifact is a valid language model at every depth." link: https://arxiv.org/abs/2609.3576911d
Danielle Fong 🔆@DanielleFongRT @iScienceLuvr: Telescopic Language Models "We train a Telescopic Language Model (TLM) to be that continuum: a nested-capacity Transform…11d
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    Tanishq Mathew Abraham, Ph.D.

    2 Sources

    Tanishq Mathew Abraham, Ph.D.@iScienceLuvrTelescopic Language Models "We train a Telescopic Language Model (TLM) to be that continuum: a nested-capacity Transformer supervised by stochastic prefix supervision with a full anchor. At every step, one randomly truncated prefix of the capacity axis is trained against the full next-token target, alongside one full-capacity pass, so the trained artifact is a valid language model at every depth." link: https://arxiv.org/abs/2609.3576911d
    Danielle Fong 🔆@DanielleFongRT @iScienceLuvr: Telescopic Language Models "We train a Telescopic Language Model (TLM) to be that continuum: a nested-capacity Transform…11d
    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet