Cross-model Transferability among Large Language Models on the Platonic Representations of Concepts

• URL: http://arxiv.org/abs/2501.02009v1
• Date: Thu, 02 Jan 2025 11:56:59 GMT
• Title: Cross-model Transferability among Large Language Models on the Platonic Representations of Concepts
• Authors: Youcheng Huang, Chen Huang, Duanyu Feng, Wenqiang Lei, Jiancheng Lv,
• Abstract summary: This work explores the intricate relationships between concept representations across different Large Language Models (LLMs)<n>We introduce a linear transformation method to bridge these representations, drawing an intriguing parallel to Plato's Allegory of the Cave.<n>A weak-to-strong transferability exists between LLM concept representations, whereby SVs extracted from smaller LLMs can effectively control the behavior of larger LLMs.
• Score: 24.990218468155383
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Understanding the inner workings of Large Language Models (LLMs) is a critical research frontier. Prior research has shown that a single LLM's concept representations can be captured as steering vectors (SVs), enabling the control of LLM behavior (e.g., towards generating harmful content). Our work takes a novel approach by exploring the intricate relationships between concept representations across different LLMs, drawing an intriguing parallel to Plato's Allegory of the Cave. In particular, we introduce a linear transformation method to bridge these representations and present three key findings: 1) Concept representations across different LLMs can be effectively aligned using simple linear transformations, enabling efficient cross-model transfer and behavioral control via SVs. 2) This linear transformation generalizes across concepts, facilitating alignment and control of SVs representing different concepts across LLMs. 3) A weak-to-strong transferability exists between LLM concept representations, whereby SVs extracted from smaller LLMs can effectively control the behavior of larger LLMs.

Related papers

• Distilling Transitional Pattern to Large Language Models for Multimodal Session-based Recommendation [67.84581846180458]
  Session-based recommendation (SBR) predicts the next item based on anonymous sessions. Recent Multimodal SBR methods utilize simplistic pre-trained models for modality learning but have limitations in semantic richness. We propose a multimodal LLM-enhanced framework TPAD, which extends a distillation paradigm to decouple and align transitional patterns for promoting MSBR.
  arXiv  Detail & Related papers  (2025-04-13T07:49:08Z)
• Toward a Flexible Framework for Linear Representation Hypothesis Using Maximum Likelihood Estimation [3.515066520628763]
  We introduce a new notion of binary concepts as unit vectors in a canonical representation space. Our method, Sum of Activation-base Normalized Difference (SAND), formalizes the use of activation differences modeled as samples from a von Mises-Fisher distribution.
  arXiv  Detail & Related papers  (2025-02-22T23:56:30Z)
• Analyzing Fine-tuning Representation Shift for Multimodal LLMs Steering alignment [53.90425382758605]
  We show how fine-tuning alters the internal structure of a model to specialize in new multimodal tasks. Our work sheds light on how multimodal representations evolve through fine-tuning and offers a new perspective for interpreting model adaptation in multimodal tasks.
  arXiv  Detail & Related papers  (2025-01-06T13:37:13Z)
• Provably Transformers Harness Multi-Concept Word Semantics for Efficient In-Context Learning [53.685764040547625]
  Transformer-based large language models (LLMs) have displayed remarkable creative prowess and emergence capabilities. This work provides a fine mathematical analysis to show how transformers leverage the multi-concept semantics of words to enable powerful ICL and excellent out-of-distribution ICL abilities.
  arXiv  Detail & Related papers  (2024-11-04T15:54:32Z)
• LLaVA-KD: A Framework of Distilling Multimodal Large Language Models [70.19607283302712]
  We propose a novel framework to transfer knowledge from l-MLLM to s-MLLM. Specifically, we introduce Multimodal Distillation (MDist) to minimize the divergence between the visual-textual output distributions of l-MLLM and s-MLLM. We also propose a three-stage training scheme to fully exploit the potential of s-MLLM.
  arXiv  Detail & Related papers  (2024-10-21T17:41:28Z)
• ConTrans: Weak-to-Strong Alignment Engineering via Concept Transplantation [35.95755927728259]
  ConTrans is a novel framework that enables weak-to-strong alignment transfer via concept transplantation.<n>Experiment results validate the effectiveness of both inter-LLM-family and intra-LLM-family concept transplantation.
  arXiv  Detail & Related papers  (2024-05-22T12:15:52Z)
• Bridging Vision and Language Spaces with Assignment Prediction [47.04855334955006]
  VLAP is a novel approach that bridges pretrained vision models and large language models (LLMs) We harness well-established word embeddings to bridge two modality embedding spaces. VLAP achieves substantial improvements over the previous linear transformation-based approaches.
  arXiv  Detail & Related papers  (2024-04-15T10:04:15Z)
• Bias Amplification in Language Model Evolution: An Iterated Learning Perspective [27.63295869974611]
  We draw parallels between the behavior of Large Language Models (LLMs) and the evolution of human culture. Our approach involves leveraging Iterated Learning (IL), a Bayesian framework that elucidates how subtle biases are magnified during human cultural evolution. This paper outlines key characteristics of agents' behavior in the Bayesian-IL framework, including predictions that are supported by experimental verification.
  arXiv  Detail & Related papers  (2024-04-04T02:01:25Z)
• Sparsity-Guided Holistic Explanation for LLMs with Interpretable Inference-Time Intervention [53.896974148579346]
  Large Language Models (LLMs) have achieved unprecedented breakthroughs in various natural language processing domains. The enigmatic black-box'' nature of LLMs remains a significant challenge for interpretability, hampering transparent and accountable applications. We propose a novel methodology anchored in sparsity-guided techniques, aiming to provide a holistic interpretation of LLMs.
  arXiv  Detail & Related papers  (2023-12-22T19:55:58Z)
• LION : Empowering Multimodal Large Language Model with Dual-Level Visual Knowledge [58.82222646803248]
  Multimodal Large Language Models (MLLMs) have endowed LLMs with the ability to perceive and understand multi-modal signals. Most of the existing MLLMs mainly adopt vision encoders pretrained on coarsely aligned image-text pairs, leading to insufficient extraction and reasoning of visual knowledge. We propose a dual-Level vIsual knedgeOwl eNhanced Multimodal Large Language Model (LION), which empowers the MLLM by injecting visual knowledge in two levels.
  arXiv  Detail & Related papers  (2023-11-20T15:56:44Z)
• Identifying Linear Relational Concepts in Large Language Models [16.917379272022064]
  Transformer language models (LMs) have been shown to represent concepts as directions in the latent space of hidden activations. We present a technique called linear relational concepts (LRC) for finding concept directions corresponding to human-interpretable concepts.
  arXiv  Detail & Related papers  (2023-11-15T14:01:41Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.