Pith. sign in

REVIEW 15 cited by

Generalization in Generation: A closer look at Exposure Bias

Not yet reviewed by Pith; the record is open.

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 1910.00292 v2 pith:J6F62ZFV submitted 2019-10-01 cs.LG cs.CLstat.ML

Generalization in Generation: A closer look at Exposure Bias

classification cs.LG cs.CLstat.ML
keywords generalizationmodelbiascontextsexposuregeneratedgenerationlearning
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Exposure bias refers to the train-test discrepancy that seemingly arises when an autoregressive generative model uses only ground-truth contexts at training time but generated ones at test time. We separate the contributions of the model and the learning framework to clarify the debate on consequences and review proposed counter-measures. In this light, we argue that generalization is the underlying property to address and propose unconditional generation as its fundamental benchmark. Finally, we combine latent variable modeling with a recent formulation of exploration in reinforcement learning to obtain a rigorous handling of true and generated contexts. Results on language modeling and variational sentence auto-encoding confirm the model's generalization capability.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 15 Pith papers

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

  1. CATCH-ME if you RAG: a dataset of Contextually Annotated multi-Turn Counterspeech against Hate and Misinformation Exchanges

    cs.CL 2026-06 unverdicted novelty 8.0

    Presents a new expert-curated dataset of multi-turn counterspeech dialogues in five languages targeting hate against seven groups, with span annotations linking to verified external knowledge for RAG applications.

  2. SDFlow: Similarity-Driven Flow Matching for Time Series Generation

    cs.AI 2026-05 unverdicted novelty 7.0

    SDFlow uses similarity-driven flow matching with low-rank manifold decomposition and a categorical posterior to generate high-fidelity long time series in VQ space without step-wise error accumulation.

  3. Learning to summarize from human feedback

    cs.CL 2020-09 conditional novelty 7.0

    Reinforcement learning on a reward model trained from human summary comparisons produces summaries humans prefer over supervised fine-tuning or human references on TL;DR and transfers to CNN/DM.

  4. DSA: Dynamic Step Allocation for Fast Autoregressive Video Generation

    cs.CV 2026-06 unverdicted novelty 6.0

    DSA adds a jointly trained confidence head to autoregressive video diffusion models that dynamically allocates fewer or more denoising steps per frame, achieving 22.63 FPS real-time generation on H100 while matching V...

  5. Flow map learning in nonlinear vector autoregressive models: influence of the feature-library structure on the training error

    cs.LG 2026-05 unverdicted novelty 6.0

    NVAR models exhibit training error scaling laws tied to feature library representation of Lie-series coefficients, with delays reducing one-step error but aiding long-horizon forecasts only under sufficient nonlinearity.

  6. Assisted Counterspeech Writing at the Crossroads of Hate Speech and Misinformation

    cs.CL 2026-05 conditional novelty 6.0

    LLMs generate adequate counterspeech for co-occurring hate and misinformation in 40% of cases, with a mixed knowledge strategy from fact-checkers and NGOs proving most effective after expert revision.

  7. Mechanisms of Misgeneralization in Physical Sequence Modeling

    cs.LG 2026-05 unverdicted novelty 6.0

    Generative sequence models for physical tasks exhibit physical misgeneralization where local prediction errors propagate through physical measurements to distort aggregate distributions over quantities like distance o...

  8. SDFlow: Similarity-Driven Flow Matching for Time Series Generation

    cs.AI 2026-05 unverdicted novelty 6.0

    SDFlow learns a global transport map via similarity-driven flow matching in VQ latent space, using low-rank manifold decomposition and a categorical posterior to handle discreteness, yielding SOTA long-horizon perform...

  9. Rolling Sink: Bridging Limited-Horizon Training and Open-Ended Testing in Autoregressive Video Diffusion

    cs.CV 2026-02 unverdicted novelty 6.0

    Rolling Sink is a training-free cache adjustment technique that maintains visual consistency in autoregressive video diffusion models for ultra-long open-ended generation beyond training horizons.

  10. MixFlow Training: Alleviating Exposure Bias with Slowed Interpolation Mixture

    cs.CV 2025-12 conditional novelty 6.0

    Training diffusion models on a mixture of higher-noise interpolations (MixFlow) improves generation FID across SiT, REPA, RAE and SD3.5, reaching ImageNet 256 gFID 1.43 after post-training.

  11. End-to-End Training for Autoregressive Video Diffusion via Self-Resampling

    cs.CV 2025-12 conditional novelty 6.0

    Resampling Forcing trains autoregressive video diffusion models on self-resampled degraded histories with a causal mask, achieving stable long-horizon generation without a teacher or discriminator.

  12. STARCaster: Spatio-Temporal AutoRegressive Video Diffusion for Identity- and View-Aware Talking Portraits

    cs.CV 2025-12 conditional novelty 6.0

    STARCaster is a 2D spatio-temporal video diffusion model that unifies identity-conditioned audio-driven portrait animation and novel-view synthesis without explicit 3D reconstruction.

  13. Rolling Forcing: Autoregressive Long Video Diffusion in Real Time

    cs.CV 2025-09 unverdicted novelty 6.0

    Rolling Forcing generates multi-minute videos in real time by jointly denoising frames at increasing noise levels, anchoring attention to early frames, and using windowed distillation to limit error accumulation.

  14. MT-EditFlow: Reinforcement Learning for Multi-Turn Image Editing with Flow Matching

    cs.CV 2026-06 unverdicted novelty 5.0

    MT-EditFlow applies flow-matching RL with multi-reward aggregation to improve multi-turn image editing performance on models like FLUX.1-Kontext-dev by 6.85 points at turn-3.

  15. GenTSE: Enhancing Target Speaker Extraction via a Coarse-to-Fine Generative Language Model

    eess.AS 2025-12 conditional novelty 5.0

    A two-stage decoder-only language model with continuous embeddings and UTMOS-based preference fine-tuning reports improved target-speaker-extraction scores on Libri2Mix.