Close Menu
techwavedispatch.com

    Subscribe to Updates

    Get the latest creative news from FooBar about art, design and business.

    What's Hot

    Learning from Context: Leave-One-Out Prediction in Jump Diffusions

    November 22, 2026

    Represented but misread: Post-training changes how language models read out what they represent

    November 22, 2026

    Mamba Flow Matching Neural Processes: Linear-Time Inference for Irregularly Observed Spatial Fields

    November 22, 2026
    Facebook X (Twitter) Instagram
    Trending
    • Learning from Context: Leave-One-Out Prediction in Jump Diffusions
    • Represented but misread: Post-training changes how language models read out what they represent
    • Mamba Flow Matching Neural Processes: Linear-Time Inference for Irregularly Observed Spatial Fields
    • Cost-Aware Agentic Feature Optimization System for Fleet-Scale Ranking Models
    • TRCF: Temporal-Reference Complementary Forecasting for Irregular Multivariate Time Series
    • [2611.26515] Simple Private Noisy SGD is Minimax-Optimal under Population Strong Convexity
    • OceanCoder: A Verification-Guided Large Language Model for Automated Simulation Modelling in Ocean Engineering
    • Bayesian optimization on amortized search space
    Facebook X (Twitter) Instagram
    techwavedispatch.com
    Subscribe
    Sunday, November 22
    • Home
    • AI Tools & Software
    • Machine Learning
    • Generative AI
    • Robotics & Automation
    • Future Tech & Hardware
    • AI in Business
    • Ethics & Regulations
    • More
      • Gadgets & Smart Devices
      • Cybersecurity & AI
      • Tech Breakthroughs
    techwavedispatch.com
    Home»Machine Learning»Represented but misread: Post-training changes how language models read out what they represent
    Machine Learning

    Represented but misread: Post-training changes how language models read out what they represent

    adminBy adminNovember 22, 2026No Comments2 Mins Read
    Facebook Twitter Pinterest LinkedIn Tumblr Email
    Represented but misread: Post-training changes how language models read out what they represent
    Share
    Facebook Twitter LinkedIn Pinterest Email

    arXiv:2611.26584v1 Announce Type: cross
    Abstract: Does post-training change what a language model represents, or how this representation is read out into an answer? We compared linear probes with the models’ own answers, analyzed with signal detection theory, in six open-weight base models and their post-trained versions. For judgments of whether German texts were written by a human or an AI, probes and word n-gram classifiers separated the classes well, but the answers did not and often ordered the texts inversely. Post-training lowered the sensitivity of the answers on average ($\Delta d’ = -0.34$ across model families; -0.29 across three prompts). It also changed the use of a stated base rate: Base models shifted their log-odds by less than the stated change in prior odds, most post-trained models by more. In two models with stage-wise checkpoints, the change occurred during supervised fine-tuning, model stitching implicated the early layers, and from mid-depth on internal states at positions after the text, not those of the text, transferred the change between checkpoints. Our corpora cannot be shared, so we add a case reproducible from public data, the truth of negated statements. Here, post-training acted in the opposite direction: Probes decoded truth at or near ceiling in every model, the answers of several base models did not use it, and post-training repaired them; by mid-depth the repair resided at the positions after the statement; below it, the representation of the statement contributed, strongly in one of four models. In a training study, adding 2,000 correctly labeled authorship judgments of news texts to otherwise identical fine-tuning data improved the answers on our newspaper and textbook corpora and inverted them on the chats, where they followed a news-trained probe of the base model. Post-training thus changed how a judgment is read out of what a model represents, and a small training signal can set its direction.

    Language misread Models PostTraining read represent Represented
    Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
    admin
    • Website

    Related Posts

    Learning from Context: Leave-One-Out Prediction in Jump Diffusions

    November 22, 2026

    Mamba Flow Matching Neural Processes: Linear-Time Inference for Irregularly Observed Spatial Fields

    November 22, 2026

    Cost-Aware Agentic Feature Optimization System for Fleet-Scale Ranking Models

    November 22, 2026
    Leave A Reply Cancel Reply

    Top Posts

    NVIDIA Is Investigating Reports That CEO Jensen Huang Was The Target Of Chinese Spy Efforts

    November 8, 202619 Views

    ADIndex: Layer-Adaptive Sparse Attention Indexing via Group-Level Distillation

    November 11, 20266 Views

    Anthropic releases new Mythos and Fable models amid RSI concerns

    November 9, 20266 Views

    One Policy for 100 Household Tasks: Our Solution for the 2026 BEHAVIOR Challenge

    November 11, 20265 Views
    Don't Miss
    Machine Learning

    Learning from Context: Leave-One-Out Prediction in Jump Diffusions

    By adminNovember 22, 20260

    arXiv:2611.26359v1 Announce Type: new Abstract: Continuous diffusion models generate discrete data through stochastic dynamics in…

    Represented but misread: Post-training changes how language models read out what they represent

    November 22, 2026

    Mamba Flow Matching Neural Processes: Linear-Time Inference for Irregularly Observed Spatial Fields

    November 22, 2026

    Cost-Aware Agentic Feature Optimization System for Fleet-Scale Ranking Models

    November 22, 2026
    About Us

    Welcome to TechWave Dispatch

    At TechWave Dispatch, we bring you the latest news, insights, and developments from the ever-evolving world of technology. Our mission is to keep readers informed about the innovations shaping our digital future.

    From artificial intelligence and machine learning to cybersecurity, robotics, smart devices, and groundbreaking technologies, we cover the topics that matter in today's fast-moving tech landscape.

    Our Picks

    Learning from Context: Leave-One-Out Prediction in Jump Diffusions

    November 22, 2026

    Represented but misread: Post-training changes how language models read out what they represent

    November 22, 2026

    Mamba Flow Matching Neural Processes: Linear-Time Inference for Irregularly Observed Spatial Fields

    November 22, 2026
    Most Popular

    After blocking Muse, Amazon now lets users text Alexa to help them shop

    November 6, 20260 Views

    A controlled world for robot learning and evaluation

    November 6, 20260 Views

    Egyptian and Nigerian English join the Open ASR Leaderboard

    November 6, 20260 Views
    Facebook X (Twitter) Instagram Pinterest
    • Terms and Conditions
    • Privacy Policy
    • Contact Us
    • About Us
    © 2026 TechWave Dispatch. All Rights Reserved.

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