Representation Engineering for Large-Language Models: Survey and Research Challenges

• URL: http://arxiv.org/abs/2502.17601v1
• Date: Mon, 24 Feb 2025 19:36:26 GMT
• Title: Representation Engineering for Large-Language Models: Survey and Research Challenges
• Authors: Lukasz Bartoszcze, Sarthak Munshi, Bryan Sukidi, Jennifer Yen, Zejia Yang, David Williams-King, Linh Le, Kosi Asuzu, Carsten Maple,
• Abstract summary: We formalize the goals and methods of representation engineering to present a cohesive picture of work in this emerging discipline.<n>We outline risks such as performance decrease, compute time increases and steerability issues.
• Score: 6.670092691529475
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large-language models are capable of completing a variety of tasks, but remain unpredictable and intractable. Representation engineering seeks to resolve this problem through a new approach utilizing samples of contrasting inputs to detect and edit high-level representations of concepts such as honesty, harmfulness or power-seeking. We formalize the goals and methods of representation engineering to present a cohesive picture of work in this emerging discipline. We compare it with alternative approaches, such as mechanistic interpretability, prompt-engineering and fine-tuning. We outline risks such as performance decrease, compute time increases and steerability issues. We present a clear agenda for future research to build predictable, dynamic, safe and personalizable LLMs.

Related papers

• Why Reasoning Matters? A Survey of Advancements in Multimodal Reasoning (v1) [66.51642638034822]
  Reasoning is central to human intelligence, enabling structured problem-solving across diverse tasks. Recent advances in large language models (LLMs) have greatly enhanced their reasoning abilities in arithmetic, commonsense, and symbolic domains. This paper offers a concise yet insightful overview of reasoning techniques in both textual and multimodal LLMs.
  arXiv  Detail & Related papers  (2025-04-04T04:04:56Z)
• Causality can systematically address the monsters under the bench(marks) [64.36592889550431]
  Benchmarks are plagued by various biases, artifacts, or leakage.<n>Models may behave unreliably due to poorly explored failure modes.<n> causality offers an ideal framework to systematically address these challenges.
  arXiv  Detail & Related papers  (2025-02-07T17:01:37Z)
• Failures Are Fated, But Can Be Faded: Characterizing and Mitigating Unwanted Behaviors in Large-Scale Vision and Language Models [7.736445799116692]
  In large deep neural networks that seem to perform surprisingly well on many tasks, we also observe a few failures related to accuracy, social biases, and alignment with human values. We introduce a post-hoc method that utilizes emphdeep reinforcement learning to explore and construct the landscape of failure modes in pre-trained discriminative and generative models. We empirically show the effectiveness of the proposed method across common Computer Vision, Natural Language Processing, and Vision-Language tasks.
  arXiv  Detail & Related papers  (2024-06-11T10:45:41Z)
• Information Theoretic Text-to-Image Alignment [49.396917351264655]
  Mutual Information (MI) is used to guide model alignment.<n>Our method uses self-supervised fine-tuning and relies on a point-wise (MI) estimation between prompts and images.<n>Our analysis indicates that our method is superior to the state-of-the-art, yet it only requires the pre-trained denoising network of the T2I model itself to estimate MI.
  arXiv  Detail & Related papers  (2024-05-31T12:20:02Z)
• Tradeoffs Between Alignment and Helpfulness in Language Models with Representation Engineering [15.471566708181824]
  We study the tradeoff between the increase in alignment and decrease in helpfulness of the model. Under the conditions of our framework, alignment can be guaranteed with representation engineering. We show that helpfulness is harmed quadratically with the norm of the representation engineering vector.
  arXiv  Detail & Related papers  (2024-01-29T17:38:14Z)
• Towards Truly Zero-shot Compositional Visual Reasoning with LLMs as Programmers [54.83459025465947]
  Even the largest models struggle with compositional reasoning, generalization, fine-grained spatial and temporal reasoning, and counting. Visual reasoning with large language models (LLMs) as controllers can, in principle, address these limitations by decomposing the task and solving subtasks by orchestrating a set of (visual) tools. We present a framework that mitigates these issues by introducing spatially and temporally abstract routines and by leveraging a small number of labeled examples to automatically generate in-context examples.
  arXiv  Detail & Related papers  (2024-01-03T20:48:47Z)
• Improving Factuality and Reasoning in Language Models through Multiagent Debate [95.10641301155232]
  We present a complementary approach to improve language responses where multiple language model instances propose and debate their individual responses and reasoning processes over multiple rounds to arrive at a common final answer. Our findings indicate that this approach significantly enhances mathematical and strategic reasoning across a number of tasks. Our approach may be directly applied to existing black-box models and uses identical procedure and prompts for all tasks we investigate.
  arXiv  Detail & Related papers  (2023-05-23T17:55:11Z)
• Task Formulation Matters When Learning Continually: A Case Study in Visual Question Answering [58.82325933356066]
  Continual learning aims to train a model incrementally on a sequence of tasks without forgetting previous knowledge. We present a detailed study of how different settings affect performance for Visual Question Answering.
  arXiv  Detail & Related papers  (2022-09-30T19:12:58Z)
• Core Challenges in Embodied Vision-Language Planning [9.190245973578698]
  We discuss Embodied Vision-Language Planning tasks, a family of prominent embodied navigation and manipulation problems. We propose a taxonomy to unify these tasks and provide an analysis and comparison of the new and current algorithmic approaches. We advocate for task construction that enables model generalizability and furthers real-world deployment.
  arXiv  Detail & Related papers  (2021-06-26T05:18:58Z)
• Which Model to Transfer? Finding the Needle in the Growing Haystack [27.660318887140203]
  We provide a formalization of this problem through a familiar notion of regret. We show that both task-agnostic and task-aware methods can yield high regret. We then propose a simple and efficient hybrid search strategy which outperforms the existing approaches.
  arXiv  Detail & Related papers  (2020-10-13T14:00:22Z)
• Goal-Aware Prediction: Learning to Model What Matters [105.43098326577434]
  One of the fundamental challenges in using a learned forward dynamics model is the mismatch between the objective of the learned model and that of the downstream planner or policy. We propose to direct prediction towards task relevant information, enabling the model to be aware of the current task and encouraging it to only model relevant quantities of the state space. We find that our method more effectively models the relevant parts of the scene conditioned on the goal, and as a result outperforms standard task-agnostic dynamics models and model-free reinforcement learning.
  arXiv  Detail & Related papers  (2020-07-14T16:42:59Z)

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.