arXiv Machine Learning July 15, 2026 By Gunner Levi Howe

What Makes a Representational Prior Work? Feature Families, Label-Free Invariances, and Critical Windows in Grokking

Read the original on arXiv Machine Learning →

arXiv:2607. 12735v1 Announce Type: new Abstract: Companion work showed the grokking delay is causally the time to form task-structured representations, injectable via a contrastive prior.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

safety

arXiv Machine Learning

Jul 7

Structure-Specific Representational Priors Causally Control the Grokking Delay

arXiv:2607. 04333v1 Announce Type: new Abstract: Grokking -- generalization arriving long after training-set interpolation -- can be accelerated by structure-agnostic interventions: gradient filtering, weight-norm clamping, geometric penalties on hidden representations.

By Gunner Levi Howe

llms

arXiv Computer Vision

Aug 24

When does fusing hand-crafted knowledge with learned representations pay? A cost-normalized benchmark of stacking, substitution, and interference

arXiv:2608.21098v1 Announce Type: new Abstract: Fusing prior knowledge with data-driven learning is attractive where data is scarce, yet no controlled account says when it helps, is redundant, or har...

By Ahmad AlMughrabi, Albert Clop, Benjamin Busam, Ricardo Marques, Petia Radeva

computer-vision benchmarks

arXiv Machine Learning

Sep 1

Ceiling-Clipped Acceptance Histograms Indicate Stranded Speed-up in Block-Diffusion Speculative Decoding

arXiv:2608.30427v1 Announce Type: cross Abstract: Speculative decoding speeds up generation with an efficient draft model (drafter) that proposes tokens for a target model to verify in one pass, pres...

By Ephrem Wu

diffusion fine-tuning benchmarks

arXiv AI

Jul 22

Circuit Claims Depend on What Is Extracted and How It Is Compared

arXiv:2607. 18921v1 Announce Type: cross Abstract: Circuit extraction identifies a small set of model components whose presence preserves a target behavior under ablation, and the resulting circuit is often read as the mechanism behind that behavior.

By Yang Sheng, Jie Fu

llms reinforcement-learning efficiency benchmarks

arXiv Computation and Language

Aug 24

Directional Contextual Representations for Dependency Relations: Why Cross-Direction Pairing Fails

arXiv:2608.20647v1 Announce Type: new Abstract: Splitting a bidirectional LSTM's contextual representation into a forward-only $F_i$ (strictly a function of tokens $1..i$) and a backward-only $B_i$ (...

By Sai Krishna Arthanari, JaeHyeong Chang, Chengzhe Sun, Siwei Lyu

arXiv Machine Learning

Aug 7

PoolBench: A Benchmark for Pooling Strategies in Concept Representation Evaluation for Decoder-Only LLMs

arXiv:2608. 05162v1 Announce Type: cross Abstract: Pooling is a consequential but under-examined design choice in decoder-only concept representation work: practitioners must collapse token-level hidden states into a passage-level vector, yet no shared protocol exists for comparing this choice across concepts, models, and tasks.

By Ayushi Agarwal

llms benchmarks

One email a morning, machine-written

Related stories

Structure-Specific Representational Priors Causally Control the Grokking Delay

When does fusing hand-crafted knowledge with learned representations pay? A cost-normalized benchmark of stacking, substitution, and interference

Ceiling-Clipped Acceptance Histograms Indicate Stranded Speed-up in Block-Diffusion Speculative Decoding

Circuit Claims Depend on What Is Extracted and How It Is Compared

Directional Contextual Representations for Dependency Relations: Why Cross-Direction Pairing Fails

PoolBench: A Benchmark for Pooling Strategies in Concept Representation Evaluation for Decoder-Only LLMs