Manipulating Transformer-Based Models: Controllability, Steerability, and Robust Interventions

• URL: http://arxiv.org/abs/2509.04549v1
• Date: Thu, 04 Sep 2025 17:32:56 GMT
• Title: Manipulating Transformer-Based Models: Controllability, Steerability, and Robust Interventions
• Abstract summary: Transformer-based language models excel in NLP tasks, but fine-grained control remains challenging.<n>This paper explores methods for manipulating transformer models through principled interventions at three levels: prompts, activations, and weights.<n>We introduce a unified framework encompassing prompt-level steering, activation interventions, and weight-space edits.
• Score: 0.2864713389096699
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Transformer-based language models excel in NLP tasks, but fine-grained control remains challenging. This paper explores methods for manipulating transformer models through principled interventions at three levels: prompts, activations, and weights. We formalize controllable text generation as an optimization problem addressable via prompt engineering, parameter-efficient fine-tuning, model editing, and reinforcement learning. We introduce a unified framework encompassing prompt-level steering, activation interventions, and weight-space edits. We analyze robustness and safety implications, including adversarial attacks and alignment mitigations. Theoretically, we show minimal weight updates can achieve targeted behavior changes with limited side-effects. Empirically, we demonstrate >90% success in sentiment control and factual edits while preserving base performance, though generalization-specificity trade-offs exist. We discuss ethical dual-use risks and the need for rigorous evaluation. This work lays groundwork for designing controllable and robust language models.

Related papers

• DART: Learning-Enhanced Model Predictive Control for Dual-Arm Non-Prehensile Manipulation [7.092954179436702]
  We present DART, a novel dual-arm framework for non-prehensile manipulation of objects on a tray.<n>Our pipeline is validated in simulation with objects of varying mass, geometry, and friction coefficients.<n>To the best of our knowledge, DART constitutes the first framework for non-prehensile dual-arm manipulation of objects on a tray.
  arXiv  Detail & Related papers  (2026-04-20T05:37:03Z)
• Steering Language Models Before They Speak: Logit-Level Interventions [9.055997973281919]
  We propose a training-free inference-time logit intervention for controllable generation.<n>Our results show that statistically grounded logit steering can achieve large, consistent, and multi-task control gains.
  arXiv  Detail & Related papers  (2026-01-16T03:00:33Z)
• RISER: Orchestrating Latent Reasoning Skills for Adaptive Activation Steering [62.63376387138257]
  We propose a plug-and-play intervention framework that adaptively steers large language models (LLMs) reasoning in activation space.<n>RISER constructs a library of reusable reasoning vectors and employs a lightweight Router to dynamically compose them for each input.<n>The Router is optimized via reinforcement learning under task-level rewards, activating latent cognitive primitives in an emergent and compositional manner.
  arXiv  Detail & Related papers  (2026-01-14T08:04:33Z)
• Attribution-Guided Decoding [24.52258081219335]
  We introduce Attribution-Guided Decoding (AGD), an interpretability-based decoding strategy.<n>Instead of directly manipulating model activations, AGD considers a set of high-probability output token candidates.<n>We demonstrate AGD's efficacy across three challenging domains.
  arXiv  Detail & Related papers  (2025-09-30T14:21:40Z)
• Model Editing as a Double-Edged Sword: Steering Agent Ethical Behavior Toward Beneficence or Harm [57.00627691433355]
  We frame agent behavior steering as a model editing task, which we term Behavior Editing.<n>We introduce BehaviorBench, a benchmark grounded in psychological moral theories.<n>We demonstrate that Behavior Editing can be used to promote ethical and benevolent behavior or, conversely, to induce harmful or malicious behavior.
  arXiv  Detail & Related papers  (2025-06-25T16:51:51Z)
• Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms [71.85633762642125]
  The vast number of parameters in models often results in highly intertwined internal representations.<n>Recent research has explored the use of sparse autoencoders (SAE) to disentangle knowledge in high-dimensional spaces for steering.<n>We propose Steering Target Atoms (STA), a novel method that isolates and manipulates disentangled knowledge components to enhance safety.
  arXiv  Detail & Related papers  (2025-05-23T17:59:18Z)
• Dynamic Manipulation of Deformable Objects in 3D: Simulation, Benchmark and Learning Strategy [88.8665000676562]
  Prior methods often simplify the problem to low-speed or 2D settings, limiting their applicability to real-world 3D tasks.<n>To mitigate data scarcity, we introduce a novel simulation framework and benchmark grounded in reduced-order dynamics.<n>We propose Dynamics Informed Diffusion Policy (DIDP), a framework that integrates imitation pretraining with physics-informed test-time adaptation.
  arXiv  Detail & Related papers  (2025-05-23T03:28:25Z)
• Towards LLM Guardrails via Sparse Representation Steering [11.710399901426873]
  Large Language Models (LLMs) have demonstrated remarkable performance in natural language generation tasks.<n>We propose a sparse encoding-based representation engineering method, named SRE, which decomposes polysemantic activations into a structured, monosemantic feature space.<n>By leveraging sparse autoencoding, our approach isolates and adjusts only task-specific sparse feature dimensions, enabling precise and interpretable steering of model behavior.
  arXiv  Detail & Related papers  (2025-03-21T04:50:25Z)
• Inference-Time Intervention in Large Language Models for Reliable Requirement Verification [2.3759432635713895]
  Inference-time intervention techniques provide a promising alternative to fine-tuning.<n>We demonstrate how interventions enable fine-grained control for automating the usually time-intensive requirement verification process.<n>Our method achieves robust and reliable outputs, significantly improving over both a baseline model and a fine-tuning approach.
  arXiv  Detail & Related papers  (2025-03-18T10:49:36Z)
• Successor Features for Efficient Multisubject Controlled Text Generation [48.37713738712319]
  We introduce SF-GEN, which is grounded in two primary concepts: successor features (SFs) and language model rectification. SF-GEN seamlessly integrates the two to enable dynamic steering of text generation with no need to alter the LLM's parameters. To the best of our knowledge, our research represents the first application of successor features in text generation.
  arXiv  Detail & Related papers  (2023-11-03T00:17:08Z)
• ConBaT: Control Barrier Transformer for Safe Policy Learning [26.023275758215423]
  Control Barrier Transformer (ConBaT) is an approach that learns safe behaviors from demonstrations in a self-supervised fashion. During deployment, we employ a lightweight online optimization to find actions that ensure future states lie within the learned safe set.
  arXiv  Detail & Related papers  (2023-03-07T20:04:28Z)
• OSCAR: Data-Driven Operational Space Control for Adaptive and Robust Robot Manipulation [50.59541802645156]
  Operational Space Control (OSC) has been used as an effective task-space controller for manipulation. We propose OSC for Adaptation and Robustness (OSCAR), a data-driven variant of OSC that compensates for modeling errors. We evaluate our method on a variety of simulated manipulation problems, and find substantial improvements over an array of controller baselines.
  arXiv  Detail & Related papers  (2021-10-02T01:21:38Z)

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.