AINeutralarXiv – CS AI · 5d ago7/10
🧠Researchers demonstrate that chain-of-thought reasoning in large language models like DeepSeek-R1 fundamentally changes how refusal mechanisms operate, requiring multi-stage interventions rather than simple activation steering. Unlike traditional LLMs where refusal exists in a single directional subspace, reasoning models jointly encode refusal across both residual activations and reasoning chains, making them more robust to direct attacks but potentially vulnerable to CoT-level manipulations.
AIBullisharXiv – CS AI · May 127/10
🧠HyperTransport is a new hypernetwork framework that dramatically accelerates activation steering for text-to-image models by amortizing optimization costs across multiple concepts. Rather than optimizing intervention parameters for each new concept (which takes minutes), the system learns to map CLIP embeddings directly to steering parameters in a single forward pass, achieving 3600-7000x speedup while matching per-concept baselines on unseen concepts.
AINeutralarXiv – CS AI · May 127/10
🧠Researchers discovered that large language models internally detect their own reasoning errors with 95% accuracy but verbally express unwarranted confidence in flawed outputs. Despite this hidden awareness, four intervention strategies failed to correct the errors, indicating the signal reflects computation quality rather than a mechanism that can be leveraged for improvement.
🧠 Llama
AINeutralarXiv – CS AI · May 97/10
🧠Researchers demonstrate that large reasoning models (LRMs) expose safety vulnerabilities in their intermediate reasoning traces that don't appear in final answers, creating a blind spot in current safety evaluation practices. Using adaptive multi-principle steering, they achieve up to 40.8% reduction in unsafe outputs while maintaining task accuracy, suggesting safety must be evaluated across the full reasoning-answer trajectory rather than just final responses.
AIBullisharXiv – CS AI · May 97/10
🧠Researchers introduce TACT, a technique using activation steering to detect and correct 'agent drift' in language model coding agents, where models either repeatedly reason over known information or issue tool calls without proper reasoning. The method improves task resolution rates by 4.8-5.8 percentage points across multiple benchmarks while reducing steps needed to complete tasks by up to 26%.
AINeutralarXiv – CS AI · May 97/10
🧠Researchers demonstrate that large language models encode social role granularity—from individual to institutional perspectives—as a structured geometric axis in their internal representations. Using activation steering, they show this axis is causally manipulable, enabling controlled shifts in response scope across different models.
🧠 Llama
AINeutralarXiv – CS AI · May 17/10
🧠Researchers systematically investigated whether Large Language Models can decouple fundamental reasoning patterns from specific problem instances by introducing reasoning conflicts between parametric knowledge and contextual instructions. The study reveals that LLMs prioritize task-appropriate reasoning over compliance with conflicting instructions, though mechanistic interventions at the activation level can steer models toward better instruction following by up to 29%.
AIBullisharXiv – CS AI · Apr 157/10
🧠Researchers introduce RePAIR, a framework enabling users to instruct large language models to forget harmful knowledge, misinformation, and personal data through natural language prompts at inference time. The system uses a training-free method called STAMP that manipulates model activations to achieve selective unlearning with minimal computational overhead, outperforming existing approaches while preserving model utility.
AIBearisharXiv – CS AI · Apr 147/10
🧠Researchers demonstrate that safety evaluations of persona-imbued large language models using only prompt-based testing are fundamentally incomplete, as activation steering reveals entirely different vulnerability profiles across model architectures. Testing across four models reveals the 'prosocial persona paradox' where conscientious personas safe under prompting become the most vulnerable to activation steering attacks, indicating that single-method safety assessments can miss critical failure modes.
🧠 Llama
AIBullisharXiv – CS AI · Mar 177/10
🧠Researchers have identified a method to control Large Language Model behavior by targeting only three specific attention heads called 'Style Modulation Heads' rather than the entire residual stream. This approach maintains model coherency while enabling precise persona and style control, offering a more efficient alternative to fine-tuning.
AINeutralarXiv – CS AI · Mar 117/10
🧠Researchers propose 'Curveball steering', a nonlinear method for controlling large language model behavior that outperforms traditional linear approaches. The study challenges the Linear Representation Hypothesis by showing that LLM activation spaces have substantial geometric distortions that require geometry-aware interventions.
AIBullisharXiv – CS AI · Mar 97/10
🧠Researchers have developed a new technique called activation steering to reduce reasoning biases in large language models, particularly the tendency to confuse content plausibility with logical validity. Their novel K-CAST method achieved up to 15% improvement in formal reasoning accuracy while maintaining robustness across different tasks and languages.
AINeutralarXiv – CS AI · 5d ago6/10
🧠Researchers present a novel method for controlling music generation in the MusicGen transformer by using activation steering techniques applied at inference time. The approach enables precise genre control through linear probes that manipulate the model's residual stream, demonstrating how interpretable AI behaviors can enhance collaborative music creation.
AINeutralarXiv – CS AI · May 96/10
🧠Researchers developed a causal probing framework to decode how Multimodal Large Language Models internally represent visual concepts, revealing that entities are encoded in localized regions while abstract concepts distribute globally across networks. The findings expose mechanistic drivers of scaling laws and uncover a disconnect between visual perception and reasoning capabilities in MLLMs.
AIBullisharXiv – CS AI · May 96/10
🧠Researchers introduce Memory Inception (MI), a training-free method for steering large language models by inserting text-derived key-value banks at selected attention layers rather than caching full prompts. MI achieves competitive control with instruction prompting while using up to 118x less storage and outperforms existing activation steering methods on personality, reasoning, and guidance tasks.
AINeutralarXiv – CS AI · Apr 146/10
🧠Researchers introduce an interactive workflow combining Sparse Autoencoders (SAE) and activation steering to make AI explainability actionable for practitioners. Through expert interviews with debugging tasks on CLIP, the study reveals that activation steering enables hypothesis testing and intervention-based debugging, though practitioners emphasize trust in observed model behavior over explanation plausibility and identify risks like ripple effects and limited generalization.
$XRP
AIBullisharXiv – CS AI · Apr 146/10
🧠Researchers introduce CoSToM, a framework that uses causal tracing and activation steering to improve Theory of Mind alignment in large language models. The work addresses a critical gap between LLMs' internal knowledge and external behavior, demonstrating that targeted interventions in specific neural layers can enhance social reasoning capabilities and dialogue quality.
AINeutralarXiv – CS AI · Apr 136/10
🧠Researchers introduce Dictionary-Aligned Concept Control (DACO), a framework that uses a curated dictionary of 15,000 multimodal concepts and Sparse Autoencoders to improve safety in multimodal large language models by steering their activations at inference time. Testing across multiple models shows DACO significantly enhances safety performance while preserving general-purpose capabilities without requiring model retraining.
AINeutralarXiv – CS AI · Apr 106/10
🧠Researchers introduce a framework for studying how emotional states affect decision-making in small language models (SLMs) used as autonomous agents. Using activation steering techniques grounded in real-world emotion-eliciting texts, they benchmark SLMs across game-theoretic scenarios and find that emotional perturbations systematically influence strategic choices, though behaviors often remain unstable and misaligned with human patterns.
AIBearisharXiv – CS AI · Apr 106/10
🧠Researchers studied how persona vectors—AI steering techniques that inject personality traits into large language models—affect educational applications like essay generation and automated grading. The study found that persona steering significantly degrades answer quality, with substantially larger negative impacts on open-ended humanities tasks compared to factual science questions, and reveals that AI scorers exhibit predictable bias patterns based on assigned personality traits.
AINeutralarXiv – CS AI · Mar 37/108
🧠New research reveals that large language models often determine their final answers before generating chain-of-thought reasoning, challenging the assumption that CoT reflects the model's actual decision process. Linear probes can predict model answers with 0.9 AUC accuracy before CoT generation, and steering these activations can flip answers in over 50% of cases.
AINeutralarXiv – CS AI · Mar 36/104
🧠Researchers investigated whether large language models can introspect by detecting perturbations to their internal states using Meta-Llama-3.1-8B-Instruct. They found that while binary detection methods from prior work were flawed due to methodological artifacts, models do show partial introspection capabilities, localizing sentence injections at 88% accuracy and discriminating injection strengths at 83% accuracy, but only for early-layer perturbations.
