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

Hybrid Reasoning Agents pitched to think in latent space and act through text

A post introducing the approach says its two modes combine latent-space thinking with text-based actions in one multi-turn flow.

James ZouJZ
Shang ZhuSZ
7 Sources, 1h ago, first seen 1h ago

TLDR

A post introducing Hybrid Reasoning Agents argues that LLM agents need not reason aloud token by token before acting. It pitches a two-mode approach that thinks in latent space and acts through text, and claims its latent thoughts are fully readable and any LLM can become a HybridAgent.

Combined views

330

7 Sources, first seen 1h ago

6 likes4 reposts

Combined views

330

7 Sources, first seen 1h ago

6 likes4 reposts
Image: @Jiaru_Zou on X

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.

7 Sources

Jiaru "Rubin" Zou@Jiaru_ZouYour LLM agent has to think out loud, token by token, before it can search, run code, or act in the world. We humans don’t. We ponder in silence and speak when it matters. Human language was never meant to be the limit of machine thought. Introducing 🌗Hybrid Reasoning Agents - Think in latent space, act through text - Two modes, one unified multi-turn flow - No black box, fully readable latent thoughts - Any LLM can become a HybridAgent1h
Pan Lu@lupantechRT @Jiaru_Zou: Your LLM agent has to think out loud, token by token, before it can search, run code, or act in the world. We humans don’t.…1h
Shang Zhu@ShangZhu18Excited to see this piece out! We built hybrid reasoning agents, that think in latent and act in text, led by the amazing @Jiaru_Zou !1h
Rohan Paul@rohanpaul_aiNew Stanford and Together AI paper finds that cheaper reasoning need not cost accuracy, as hybrid agents cut decoding compute by 62.5% to 72.9% and still scored higher. Shows that agents run faster and score higher when most thinking happens in hidden vectors, not written text. Today's agents write every reasoning step as text tokens, which is slow and costly. Models that reason in their internal vectors are faster, but they cannot call tools and humans cannot read what they think. HybridAgent adds a small adapter that feeds the model's internal state back as its next input, so it can think without writing words. It switches to text for tool calls and key decisions, and its silent steps can still be decoded into readable words. On GAIA, the 14B version scored 52.4, above 43.7 for the same model trained as a text-only agent with ARPO. It also ran 2.1× faster than matched text agents. If your agent spends tokens on routine reasoning, try keeping that reasoning internal and saving text for tool calls and steps humans must check.47m
James Zou@james_y_zouRT @ShangZhu18: Excited to see this piece out! We built hybrid reasoning agents, that think in latent and act in text, led by the amazing @…43m
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI

    7 Sources

    Jiaru "Rubin" Zou@Jiaru_ZouYour LLM agent has to think out loud, token by token, before it can search, run code, or act in the world. We humans don’t. We ponder in silence and speak when it matters. Human language was never meant to be the limit of machine thought. Introducing 🌗Hybrid Reasoning Agents - Think in latent space, act through text - Two modes, one unified multi-turn flow - No black box, fully readable latent thoughts - Any LLM can become a HybridAgent1h
    Pan Lu@lupantechRT @Jiaru_Zou: Your LLM agent has to think out loud, token by token, before it can search, run code, or act in the world. We humans don’t.…1h
    Shang Zhu@ShangZhu18Excited to see this piece out! We built hybrid reasoning agents, that think in latent and act in text, led by the amazing @Jiaru_Zou !1h
    Rohan Paul@rohanpaul_aiNew Stanford and Together AI paper finds that cheaper reasoning need not cost accuracy, as hybrid agents cut decoding compute by 62.5% to 72.9% and still scored higher. Shows that agents run faster and score higher when most thinking happens in hidden vectors, not written text. Today's agents write every reasoning step as text tokens, which is slow and costly. Models that reason in their internal vectors are faster, but they cannot call tools and humans cannot read what they think. HybridAgent adds a small adapter that feeds the model's internal state back as its next input, so it can think without writing words. It switches to text for tool calls and key decisions, and its silent steps can still be decoded into readable words. On GAIA, the 14B version scored 52.4, above 43.7 for the same model trained as a text-only agent with ARPO. It also ran 2.1× faster than matched text agents. If your agent spends tokens on routine reasoning, try keeping that reasoning internal and saving text for tool calls and steps humans must check.47m
    James Zou@james_y_zouRT @ShangZhu18: Excited to see this piece out! We built hybrid reasoning agents, that think in latent and act in text, led by the amazing @…43m
    Today's Rank

    #13

    Today's Rank

    #13