Pith. sign in

REVIEW 4 cited by

Training Plug-n-Play Knowledge Modules with Deep Context Distillation

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 2503.08727 v4 pith:3Y5E44A4 submitted 2025-03-11 cs.LG cs.AI

classification cs.LGcs.AI
keywords trainingcontextinformationknowledgemodulesdeepdistillationdocument
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Dynamically integrating new or rapidly evolving information after (Large) Language Model pre-training remains challenging, particularly in low-data scenarios or when dealing with private and specialized documents. In-context learning and retrieval-augmented generation (RAG) face limitations, including their high inference costs and their inability to capture global document information. In this paper, we propose a way of modularizing knowledge by training document-level Knowledge Modules (KMs). KMs are lightweight components implemented as parameter-efficient LoRA modules, which are trained to store information about new documents and can be easily plugged into models on demand. We show that next-token prediction performs poorly as the training objective for KMs. We instead propose Deep Context Distillation: we learn KMs parameters such as to simulate hidden states and logits of a teacher that takes the document in context. Our method outperforms standard next-token prediction and pre-instruction training techniques, across two datasets. Finally, we highlight synergies between KMs and RAG.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

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

  1. Language Models Need Sleep: Learning to Self-Modify and Consolidate Memories

    cs.LG 2026-06 unverdicted novelty 6.0 of 10

    Sleep-time Knowledge Seeding plus Dreaming lets LLMs expand capacity, distill fragile in-context memories into stable parameters, and self-improve without human labels.

  2. Adapt Once, Thrive with Updates: Transferable Parameter-Efficient Fine-Tuning on Evolving Base Models

    cs.CL 2025-06 conditional novelty 6.0 of 10

    Trans-PEFT uses random FFN masking and layer dropping during fine-tuning so PEFT modules trained on an old base model transfer to a continually updated base model without retuning.

  3. Cartridges: Lightweight and general-purpose long context representations via self-study

    cs.CL 2025-06 conditional novelty 6.0 of 10

    A per-corpus trained KV cache, called a Cartridge, matches full-context in-context learning quality on long-document benchmarks while using up to 38.6x less serving memory.

  4. LoRA-Augmented Generation (LAG) for Knowledge-Intensive Language Tasks

    cs.CL 2025-07 conditional novelty 5.0 of 10

    LAG is a two-stage router that filters a 1,000-adapter LoRA library with Arrow and reranks with SpectR, outperforming the Arrow baseline and reaching 92.1% of its Oracle's performance on KILT tasks.

Pith tools