arXiv AI October 2, 2026 By Fabio J. Fehr, Philip Torr

On-the-fly Weight Generation: A Hypernetwork Proof of Concept on ARC-1D

The paper demonstrates that a hypernetwork can generate tiny specialist models for individual transformations in the ARC-1D benchmark directly from a few demonstrations. These generated parameters form a structured weight space, enabling partial compositional generalisation and improved performance on unseen transformations when explicit task identifiers are omitted. The study provides a proof of concept that few‑shot task context can be compiled on‑the‑fly into compact executable model parameters that support reuse and generalisation beyond known functions.

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 AI.

arXiv AI

Jul 9

Cost-Effective Agent Harnesses for Abstract Reasoning and Generalization on ARC-AGI-1

arXiv:2607. 06764v1 Announce Type: new Abstract: Recent progress on ARC-AGI-1 from disclosed architectures has come broadly from two regimes: heavy test-time compute over frontier models (evolutionary search, exhaustive sampling, extended chain-of-thought), or benchmark-specific training in which small models are fine-tuned on ARC data, often with task-specialized architectures.

By Kabir Moghe, Peter Chin

agents fine-tuning benchmarks

arXiv Machine Learning

Jul 23

Scaling Laws for Hypernetwork-Based Knowledge Injection in Large Language Models

arXiv:2607. 19604v1 Announce Type: cross Abstract: Injecting factual knowledge into large language models (LLMs) reliably and at scale remains an open challenge.

By Nischay Dhankhar, Dos Baha, Abulhair Saparov

llms fine-tuning

arXiv AI

Jul 1

From Search to Synthesis: Training LLMs as Zero-Shot Workflow Generators

arXiv:2606. 30704v1 Announce Type: cross Abstract: Large language models (LLMs) excel across a wide range of tasks, yet their instance-specific solutions often lack the structural consistency needed for reliable deployment.

By Gan Luo, Zihan Qin, Bin Dong, Wotao Yin

llms nlp reinforcement-learning fine-tuning benchmarks

arXiv AI

Oct 27

In-Context Adapter Generation for Compositional Generalization in Large Audio Language Models

arXiv:2610.34413v1 Announce Type: cross Abstract: Large audio language models (LALMs) have achieved strong performance on standard speech and audio understanding tasks, yet their ability to systemati...

By Yiwei Guo, Siqi Huang, Bohan Li, Jing Peng, Hankun Wang, Zhihan Li, Kai Yu

llms rag fine-tuning

arXiv AI

Sep 21

Implicit Rule Induction with Test-Time Task Embeddings in ARC-like Tasks

The paper introduces a two‑step test‑time training protocol, Embed‑TTT, for Vision ARC (VARC) that first fine‑tunes only the task embedding and then fine‑tunes the backbone. This approach consistently produces task embeddings that better align with the underlying rules, improves retrieval and linear probing, and recovers the geometric structure of parametric rules. Even fine‑tuning only the tiny embedding component solves a significant portion of ARC‑AGI‑1, ConceptARC, and Mini‑ARC tasks, while the full two‑step pipeline further enhances performance and demonstrates compositional rule interpolation.

By Adrien Deli\`ege, Claas Beger, Marc Van Droogenbroeck, Melanie Mitchell

rag benchmarks

arXiv AI

Oct 2

Capturing In-Context Learning Dynamics with Task Operators

arXiv:2610.01054v1 Announce Type: cross Abstract: In-context learning (ICL) enables language models to perform new tasks from demonstrations without weight updates. However, every ICL inference requi...

By Guangzhi Xiong, Zhenghao He, Bohan Liu, Sanchit Sinha, Wenqian Ye, Aidong Zhang

llms

One email a morning, machine-written

Related stories

Cost-Effective Agent Harnesses for Abstract Reasoning and Generalization on ARC-AGI-1

Scaling Laws for Hypernetwork-Based Knowledge Injection in Large Language Models

From Search to Synthesis: Training LLMs as Zero-Shot Workflow Generators

In-Context Adapter Generation for Compositional Generalization in Large Audio Language Models

Implicit Rule Induction with Test-Time Task Embeddings in ARC-like Tasks

Capturing In-Context Learning Dynamics with Task Operators