Pith. sign in

REVIEW 6 cited by

Knowledge Fusion of Chat LLMs: A Preliminary Technical Report

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 2402.16107 v6 pith:VJ2KNKHP submitted 2024-02-25 cs.CL

classification cs.CL
keywords llmschatfusionknowledgetargetfine-tuningfusellmfusionchat
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Recently, FuseLLM introduced the concept of knowledge fusion to transfer the collective knowledge of multiple structurally varied LLMs into a target LLM through lightweight continual training. In this report, we extend the scalability and flexibility of the FuseLLM framework to realize the fusion of chat LLMs, resulting in FusionChat. FusionChat comprises two main stages. Firstly, we undertake knowledge fusion for structurally and scale-varied source LLMs to derive multiple target LLMs of identical structure and size via lightweight fine-tuning. Then, these target LLMs are merged within the parameter space, wherein we propose a novel method for determining the merging weights based on the variation ratio of parameter matrices before and after fine-tuning. We validate our approach using three prominent chat LLMs with diverse architectures and scales, namely NH2-Mixtral-8x7B, NH2-Solar-10.7B, and OpenChat-3.5-7B. Experimental results spanning various chat domains demonstrate the superiority of FusionChat-7B across a broad spectrum of chat LLMs at 7B and 34B scales, even surpassing GPT-3.5 (March) and approaching Mixtral-8x7B-Instruct.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 6 Pith papers

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

  1. Contrastive Private Data Synthesis via Weighted Multi-PLM Fusion

    cs.LG 2025-02 conditional novelty 6.0 of 10

    WASP fuses multiple pretrained language models with differentially private sample voting and contrastive prompts to synthesize task-specific text data, improving downstream classifier accuracy over single-model baseli...

  2. Weighted-Reward Preference Optimization for Implicit Model Fusion

    cs.CL 2024-12 conditional novelty 6.0 of 10

    WRPO tunes an 8B chat model by combining its own preferred responses (on-policy) with high-reward responses from ten heterogeneous source LLMs (off-policy) using an annealed weight, beating prior fusion and preference...

  3. Enabling Flexible Multi-LLM Integration for Scalable Knowledge Aggregation

    cs.CL 2025-05 conditional novelty 5.0 of 10

    Adaptive selection and dynamic weighted fusion of source LLMs reduces knowledge interference and improves target model accuracy compared to FuseLLM.

  4. InfiFusion: A Unified Framework for Enhanced Cross-Model Reasoning via LLM Fusion

    cs.CL 2025-01 reject novelty 5.0 of 10

    InfiFusion fuses multiple large language models into one pivot model using enhanced universal logit distillation, and reports that the fused model outperforms all source models on 11 benchmarks with a fraction of the ...

  5. SeMe: Training-Free Language Model Merging via Semantic Alignment

    cs.CL 2025-05 reject novelty 4.0 of 10

    SeMe claims a data-free, training-free layer-wise language model merging method via semantic alignment, but the paper provides no method specification and no experiment results.

  6. Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks

    cs.LG 2025-04 conditional novelty 4.0 of 10

    The paper organizes large-small model collaboration into downward, upward, and inference-time transfer, and advocates multi-objective benchmarks for private-domain tasks.

Pith tools