Close Menu
  • Home
  • AI Models
    • DeepSeek
    • xAI
    • OpenAI
    • Meta AI Llama
    • Google DeepMind
    • Amazon AWS AI
    • Microsoft AI
    • Anthropic (Claude)
    • NVIDIA AI
    • IBM WatsonX Granite 3.1
    • Adobe Sensi
    • Hugging Face
    • Alibaba Cloud (Qwen)
    • Baidu (ERNIE)
    • C3 AI
    • DataRobot
    • Mistral AI
    • Moonshot AI (Kimi)
    • Google Gemma
    • xAI
    • Stability AI
    • H20.ai
  • AI Research
    • Allen Institue for AI
    • arXiv AI
    • Berkeley AI Research
    • CMU AI
    • Google Research
    • Microsoft Research
    • Meta AI Research
    • OpenAI Research
    • Stanford HAI
    • MIT CSAIL
    • Harvard AI
  • AI Funding & Startups
    • AI Funding Database
    • CBInsights AI
    • Crunchbase AI
    • Data Robot Blog
    • TechCrunch AI
    • VentureBeat AI
    • The Information AI
    • Sifted AI
    • WIRED AI
    • Fortune AI
    • PitchBook
    • TechRepublic
    • SiliconANGLE – Big Data
    • MIT News
    • Data Robot Blog
  • Expert Insights & Videos
    • Google DeepMind
    • Lex Fridman
    • Matt Wolfe AI
    • Yannic Kilcher
    • Two Minute Papers
    • AI Explained
    • TheAIEdge
    • Matt Wolfe AI
    • The TechLead
    • Andrew Ng
    • OpenAI
  • Expert Blogs
    • François Chollet
    • Gary Marcus
    • IBM
    • Jack Clark
    • Jeremy Howard
    • Melanie Mitchell
    • Andrew Ng
    • Andrej Karpathy
    • Sebastian Ruder
    • Rachel Thomas
    • IBM
  • AI Policy & Ethics
    • ACLU AI
    • AI Now Institute
    • Center for AI Safety
    • EFF AI
    • European Commission AI
    • Partnership on AI
    • Stanford HAI Policy
    • Mozilla Foundation AI
    • Future of Life Institute
    • Center for AI Safety
    • World Economic Forum AI
  • AI Tools & Product Releases
    • AI Assistants
    • AI for Recruitment
    • AI Search
    • Coding Assistants
    • Customer Service AI
    • Image Generation
    • Video Generation
    • Writing Tools
    • AI for Recruitment
    • Voice/Audio Generation
  • Industry Applications
    • Finance AI
    • Healthcare AI
    • Legal AI
    • Manufacturing AI
    • Media & Entertainment
    • Transportation AI
    • Education AI
    • Retail AI
    • Agriculture AI
    • Energy AI
  • AI Art & Entertainment
    • AI Art News Blog
    • Artvy Blog » AI Art Blog
    • Weird Wonderful AI Art Blog
    • The Chainsaw » AI Art
    • Artvy Blog » AI Art Blog
What's Hot

VOGUE: Guiding Exploration with Visual Uncertainty Improves Multimodal Reasoning – Takara TLDR

Thinking Machines debuts Tinker, a developer tool to simplify fine-tuning of AI models | Technology News

What to expect from free Perplexity AI Comet Browser: Enhanced multitasking?

Facebook X (Twitter) Instagram
Advanced AI News
  • Home
  • AI Models
    • OpenAI (GPT-4 / GPT-4o)
    • Anthropic (Claude 3)
    • Google DeepMind (Gemini)
    • Meta (LLaMA)
    • Cohere (Command R)
    • Amazon (Titan)
    • IBM (Watsonx)
    • Inflection AI (Pi)
  • AI Research
    • Allen Institue for AI
    • arXiv AI
    • Berkeley AI Research
    • CMU AI
    • Google Research
    • Meta AI Research
    • Microsoft Research
    • OpenAI Research
    • Stanford HAI
    • MIT CSAIL
    • Harvard AI
  • AI Funding
    • AI Funding Database
    • CBInsights AI
    • Crunchbase AI
    • Data Robot Blog
    • TechCrunch AI
    • VentureBeat AI
    • The Information AI
    • Sifted AI
    • WIRED AI
    • Fortune AI
    • PitchBook
    • TechRepublic
    • SiliconANGLE – Big Data
    • MIT News
    • Data Robot Blog
  • AI Experts
    • Google DeepMind
    • Lex Fridman
    • Meta AI Llama
    • Yannic Kilcher
    • Two Minute Papers
    • AI Explained
    • TheAIEdge
    • The TechLead
    • Matt Wolfe AI
    • Andrew Ng
    • OpenAI
    • Expert Blogs
      • François Chollet
      • Gary Marcus
      • IBM
      • Jack Clark
      • Jeremy Howard
      • Melanie Mitchell
      • Andrew Ng
      • Andrej Karpathy
      • Sebastian Ruder
      • Rachel Thomas
      • IBM
  • AI Tools
    • AI Assistants
    • AI for Recruitment
    • AI Search
    • Coding Assistants
    • Customer Service AI
  • AI Policy
    • ACLU AI
    • AI Now Institute
    • Center for AI Safety
  • Business AI
    • Advanced AI News Features
    • Finance AI
    • Healthcare AI
    • Education AI
    • Energy AI
    • Legal AI
LinkedIn Instagram YouTube Threads X (Twitter)
Advanced AI News
Hugging Face

Think Right: Learning to Mitigate Under-Over Thinking via Adaptive, Attentive Compression – Takara TLDR

By Advanced AI EditorOctober 5, 2025No Comments2 Mins Read
Share Facebook Twitter Pinterest Copy Link Telegram LinkedIn Tumblr Email
Share
Facebook Twitter LinkedIn Pinterest Email


Recent thinking models solve complex reasoning tasks by scaling test-time
compute, but this scaling must be allocated in line with task difficulty. On
one hand, short reasoning (underthinking) leads to errors on harder problems
that require extended reasoning steps; but, excessively long reasoning
(overthinking) can be token-inefficient, generating unnecessary steps even
after reaching a correct intermediate solution. We refer to this as
under-adaptivity, where the model fails to modulate its response length
appropriately given problems of varying difficulty. To address under-adaptivity
and strike a balance between under- and overthinking, we propose TRAAC (Think
Right with Adaptive, Attentive Compression), an online post-training RL method
that leverages the model’s self-attention over a long reasoning trajectory to
identify important steps and prune redundant ones. TRAAC also estimates
difficulty and incorporates it into training rewards, thereby learning to
allocate reasoning budget commensurate with example difficulty. Our approach
improves accuracy, reduces reasoning steps, and enables adaptive thinking
compared to base models and other RL baselines. Across a variety of tasks
(AIME, AMC, GPQA-D, BBEH), TRAAC (Qwen3-4B) achieves an average absolute
accuracy gain of 8.4% with a relative reduction in reasoning length of 36.8%
compared to the base model, and a 7.9% accuracy gain paired with a 29.4% length
drop compared to the best RL baseline. TRAAC also shows strong generalization:
although our models are trained on math datasets, they show accuracy and
efficiency gains on out-of-distribution non-math datasets like GPQA-D, BBEH,
and OptimalThinkingBench. Our analysis further verifies that TRAAC provides
fine-grained adjustments to thinking budget based on difficulty and that a
combination of task-difficulty calibration and attention-based compression
yields gains across diverse tasks.



Source link

Follow on Google News Follow on Flipboard
Share. Facebook Twitter Pinterest LinkedIn Tumblr Email Copy Link
Previous ArticleIs Perplexity’s Comet browser the next big challenger to Chrome?
Next Article The Lean AI Lab’s Blueprint for Superhuman Productivity
Advanced AI Editor
  • Website

Related Posts

VOGUE: Guiding Exploration with Visual Uncertainty Improves Multimodal Reasoning – Takara TLDR

October 5, 2025

TimeSeriesScientist: A General-Purpose AI Agent for Time Series Analysis – Takara TLDR

October 5, 2025

VLA-R1: Enhancing Reasoning in Vision-Language-Action Models – Takara TLDR

October 5, 2025

Comments are closed.

Latest Posts

Former ARTnews Publisher Dies at 97

National Gallery of Art Closes as a Result of Government Shutdown

Almine Rech Closes London Gallery After More Than a Decade

Record Exec and Art Collector Gets Over 4 Years

Latest Posts

VOGUE: Guiding Exploration with Visual Uncertainty Improves Multimodal Reasoning – Takara TLDR

October 5, 2025

Thinking Machines debuts Tinker, a developer tool to simplify fine-tuning of AI models | Technology News

October 5, 2025

What to expect from free Perplexity AI Comet Browser: Enhanced multitasking?

October 5, 2025

Subscribe to News

Subscribe to our newsletter and never miss our latest news

Subscribe my Newsletter for New Posts & tips Let's stay updated!

Recent Posts

  • VOGUE: Guiding Exploration with Visual Uncertainty Improves Multimodal Reasoning – Takara TLDR
  • Thinking Machines debuts Tinker, a developer tool to simplify fine-tuning of AI models | Technology News
  • What to expect from free Perplexity AI Comet Browser: Enhanced multitasking?
  • TimeSeriesScientist: A General-Purpose AI Agent for Time Series Analysis – Takara TLDR
  • The Lean AI Lab’s Blueprint for Superhuman Productivity

Recent Comments

  1. Jodi Ellerbusch on Nuclear power investment is growing. These stocks offer exposure
  2. Tera Puryear on VAST Data Powers Smarter, Evolving AI Agents with NVIDIA Data Flywheel
  3. Link Alternatif BEJOGAMING on 1-800-CHAT-GPT—12 Days of OpenAI: Day 10
  4. DichaelBam on Michio Kaku: The Greatest Destroyer of Scientists is Junior High School | AI Podcast Clips
  5. Lewiszix on OpenAI countersues Elon Musk, calls for enjoinment from ‘further unlawful and unfair action’

Welcome to Advanced AI News—your ultimate destination for the latest advancements, insights, and breakthroughs in artificial intelligence.

At Advanced AI News, we are passionate about keeping you informed on the cutting edge of AI technology, from groundbreaking research to emerging startups, expert insights, and real-world applications. Our mission is to deliver high-quality, up-to-date, and insightful content that empowers AI enthusiasts, professionals, and businesses to stay ahead in this fast-evolving field.

Subscribe to Updates

Subscribe to our newsletter and never miss our latest news

Subscribe my Newsletter for New Posts & tips Let's stay updated!

LinkedIn Instagram YouTube Threads X (Twitter)
  • Home
  • About Us
  • Advertise With Us
  • Contact Us
  • DMCA
  • Privacy Policy
  • Terms & Conditions
© 2025 advancedainews. Designed by advancedainews.

Type above and press Enter to search. Press Esc to cancel.