Paper Detail

Learning What to Remember: Test-Time Training via Context Distillation

Zixuan Wang, Xingyu Dang, Rui-Jie Zhu, Zixin Wen, Hengyu Fu, Wenhao Chai, Jason D. Lee

arxiv Score 15.0

Published 2026-08-03 · First seen 2026-08-04

Research Track A · General AI

Abstract

Effective long-context modeling is not merely about retaining more of the past, but about preserving the information that may prove relevant later. Test-time training (TTT) is an appealing approach that performs online parameter updates for long-context modeling, yet existing TTT methods only optimize either reconstruction or online adaptation objectives without considering the future utility of retained information. In this work, we propose \textbf{T}est-\textbf{T}ime \textbf{C}ontext \textbf{D}istillation (TTCD), a TTT framework that introduces a self-supervised objective for allocating limited memory capacity for future use. Specifically, TTCD uses a long-window teacher to supervise the fast weights of a short-window student, where the hidden-state discrepancy between them offers a dense, self-supervised signal guiding the model to memorize the contextual information crucial for future token predictions. We focus on an in-place variant: In-Place TTCD (IP-TTCD), which uses the existing MLP parameters as the fast weights. Experiments on long-context language modeling tasks show IP-TTCD consistently outperforms DeltaNet, Gated DeltaNet, sliding-window attention, and TTT when pre-trained from scratch. Furthermore, IP-TTCD allows pre-trained transformer models to adapt their parameters during inference through continual pre-training, gaining long-context capabilities with only a lightweight architectural augmentation. Our results position TTCD as a step toward architectural continual learning.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
now
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{wang2026learning,
  title = {Learning What to Remember: Test-Time Training via Context Distillation},
  author = {Zixuan Wang and Xingyu Dang and Rui-Jie Zhu and Zixin Wen and Hengyu Fu and Wenhao Chai and Jason D. Lee},
  year = {2026},
  abstract = {Effective long-context modeling is not merely about retaining more of the past, but about preserving the information that may prove relevant later. Test-time training (TTT) is an appealing approach that performs online parameter updates for long-context modeling, yet existing TTT methods only optimize either reconstruction or online adaptation objectives without considering the future utility of retained information. In this work, we propose \textbackslash{}textbf\{T\}est-\textbackslash{}textbf\{T\}ime \textbackslash{}textbf\{C\}ontext \textbackslash{}textbf\{D},
  url = {https://arxiv.org/abs/2608.01672},
  keywords = {cs.CL, cs.AI, cs.LG},
  eprint = {2608.01672},
  archiveprefix = {arXiv},
}

Metadata

{}