Concept · 17 episode(s)

Ablation Studies

← all concepts

Definition

Ablation studies are experiments that selectively remove, disable, or replace one component of a system to measure how much that piece contributes to overall performance. They are the workhorse method for arguing causality in ML papers: if accuracy collapses when you delete a module, that module was doing real work.

Episodes covering this

  1. 079
    An Old Idea From Cognitive Psychology Reshapes How We Reward Reasoning Models
    Chen, Xu, Zhao et al. · Tongji University / Shanghai AI Laboratory / Nanyang Technological University·29 min·May 25, 2026
  2. 078
    Training a Markdown File: When LLM Self-Improvement Borrows the Discipline of Neural Net Training
    Yang, Gong, Huang et al. · Microsoft·28 min·May 25, 2026
  3. 077
    Reading a Model's Confidence Curve to Decide When Chain-of-Thought Is Worth It
    Xia, Wang, Tang et al. · State Key Laboratory of General Artificial Intelligence·22 min·May 25, 2026
  4. 076
    Same Model, Organized Differently: How an Agent Architecture Beat Frontier Systems at Research Math
    Zhao, Yuan, Choi et al. · Georgia Institute of Technology·22 min·May 25, 2026
  5. 075
    Growing Code and Proof Together: Verified Systems in Ten Hours Instead of a Year
    Agarwal, Krentsel, Liu et al. · UC Berkeley·28 min·May 25, 2026
  6. 074
    How a Fifteen-Hundred-Dollar Training Run Matched Llama and Gemma on Reasoning
    Wang, Liu, Wang et al. · Sapient Intelligence·21 min·May 24, 2026
  7. 071
    When the Model Is Fine and the Plumbing Is Broken: Fixing Agents at the Interface
    Xu, Wen, Li · Peking University·23 min·May 22, 2026
  8. 070
    When Models Know the Answer But Say the Wrong Thing Anyway
    Yeom, Sok, Kim et al. · Graduate School of Data Science·22 min·May 22, 2026
  9. 069
    When Smarter Models Forecast Worse: The Hidden Failure Mode in LLM Predictions
    Merrill, Lee, Karger · Forecasting Research Institute / UC Berkeley·30 min·May 22, 2026
  10. 064
    When Agent Memory Stops Being a Database and Starts Being a Skill
    Ye, Liu, Wang et al. · University of Illinois Urbana-Champaign·30 min·May 22, 2026
  11. 053
    An AI Agent Swapped In Focal Loss And Beat A Human-Tuned Training Script
    Pepe, Lin, Magka et al. · FAIR at Meta·32 min·May 18, 2026
  12. 041
    When the Iteration Teaches the Model to Skip the Iteration
    Fein-Ashley, Rashidinejad · University of Southern California·30 min·May 13, 2026
  13. 040
    Two Frozen Models Learn to Whisper: Coupling Through Hidden States
    Flamant, Ghai, Shimizu · AWS Agentic AI·29 min·May 13, 2026
  14. 033
    Echo: The Paper Arguing You Never Needed a KV Cache for Retrieval
    Sridhar, Johansen · California·24 min·May 11, 2026
  15. 021
    Ten Thousand Examples Beat the Full Industrial Pipeline for Search Agents
    Du, Ye, Tang et al. · Shanghai Jiao Tong University·14 min·May 06, 2026
  16. 013
    Why Search Keeps Rediscovering the Same Workflow, and What That Means
    Du, Liu, Du et al. · Carnegie Mellon University·22 min·May 03, 2026
  17. 012
    Why AI Coding Agents Keep Trying to Debug Without a Debugger
    Liu, Wang, Chen et al. · Sun Yat-sen University·21 min·May 02, 2026