Pith. sign in

REVIEW 10 cited by

Your Classifier is Secretly an Energy Based Model and You Should Treat it Like One

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1912.03263 v3 pith:QBGDF2RU submitted 2019-12-06 cs.LG cs.CVstat.ML

Will Grathwohl , Kuan-Chieh Wang , Jörn-Henrik Jacobsen , David Duvenaud , Mohammad Norouzi , Kevin Swersky This is my paper
classification cs.LGcs.CVstat.ML
keywords energymodelstandarddiscriminativetrainingapproachclassifierdistribution
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We propose to reinterpret a standard discriminative classifier of p(y|x) as an energy based model for the joint distribution p(x,y). In this setting, the standard class probabilities can be easily computed as well as unnormalized values of p(x) and p(x|y). Within this framework, standard discriminative architectures may beused and the model can also be trained on unlabeled data. We demonstrate that energy based training of the joint distribution improves calibration, robustness, andout-of-distribution detection while also enabling our models to generate samplesrivaling the quality of recent GAN approaches. We improve upon recently proposed techniques for scaling up the training of energy based models and presentan approach which adds little overhead compared to standard classification training. Our approach is the first to achieve performance rivaling the state-of-the-artin both generative and discriminative learning within one hybrid model.

Discussion (0). Sign in to comment.

Forward citations

Cited by 10 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Transferable Direct Prompt Injection via Activation-Guided MCMC Sampling

    cs.AI 2025-09 conditional novelty 7.0 of 10

    An activation-guided energy model plus MCMC sampling creates transferable direct prompt injection attacks in a black-box setting, reaching 49.6% average attack success across five LLMs.

  2. Revisiting Autoregressive Models for Generative Image Classification

    cs.CV 2026-03 accept novelty 6.5 of 10

    Order-marginalized any-order AR models (RandAR) outperform diffusion generative classifiers on ImageNet and OOD sets and match strong SSL models at far lower cost.

  3. BioKD: Selective Physiology-to-Video Knowledge Distillation via Reliability Gate for Emotion Recognition

    cs.LG 2026-08 conditional novelty 6.0 of 10

    BioKD uses a reliability gate and progressive distillation so that noisy physiological signals improve a video-only emotion recognition model, beating baselines on DEAP and AMIGOS.

  4. ReFP-AD: Rectified Flow Preconditioning for Energy-Based Anomaly Detection

    cs.LG 2026-08 conditional novelty 6.0 of 10

    ReFP-AD uses rectified-flow preconditioning to make finite-step MCMC stable for energy-based anomaly detection on full-dimensional DINOv2 tokens, achieving strong AUROC on MVTec-AD and VisA.

  5. Multi-Turn On-Policy Distillation with Prefix Replay

    cs.LG 2026-07 conditional novelty 6.0 of 10

    ReOPD offline-distills multi-turn agentic LLMs via teacher-prefix replay plus step-decay sampling, matching online OPD accuracy at ≥4× speed with zero tool calls.

  6. Minimal-Dissipation Learning for Energy-Based Models

    cond-mat.stat-mech 2025-10 conditional novelty 6.0 of 10

    For a harmonic-trap EBM, a finite-time learning-rate schedule trains the model to its target with provably minimal energy dissipation; its general-potential analogue is a natural-gradient flow.

  7. Silence is Golden: Leveraging Adversarial Examples to Nullify Audio Control in LDM-based Talking-Head Generation

    cs.GR 2025-06 conditional novelty 6.0 of 10

    Silencer adds a nearly invisible disturbance to portraits that makes LDM-based talking-head models keep the mouth silent, and it survives several image-purification countermeasures.

  8. BindEnergyCraft: Casting Protein Structure Predictors as Energy-Based Models for Binder Design

    cs.LG 2025-05 conditional novelty 6.0 of 10

    pTMEnergy converts AlphaFold pAE confidence logits into an energy-like score that improves computational binder design success and virtual screening over ipTM-based and generative baselines.

  9. Energy Landscapes Enable Reliable Abstention in Retrieval-Augmented Large Language Models for Healthcare

    cs.CL 2025-08 conditional novelty 5.0 of 10

    An energy-based scoring head trained on dense embeddings improves abstention decisions for medical RAG systems on semantically hard out-of-distribution queries compared to softmax and kNN baselines.

  10. Beyond Equilibrium: Non-Equilibrium Foundations Should Underpin Generative Processes in Complex Dynamical Systems

    cs.CE 2025-05 conditional novelty 3.0 of 10

    A position paper arguing that non-equilibrium-physics-inspired generative models (like diffusion models) are, and should be, the foundation for modeling time-varying complex systems, supported by one 2D simulation.

Pith tools