Language Model Steering

Large language models exhibit strong multilingual capabilities, yet significant performance gaps persist between dominant and non-dominant languages. Prior work attributes this gap to imbalances between shared and language-specific neurons in multilingual representations. We propose Cross-Lingual Activation Steering (CLAS), a training-free inference-time intervention that selectively modulates neuron activations. We evaluate CLAS on classification and generation benchmarks, achieving average improvements of 2.3% (Acc.) and 3.4% (F1) respectively, while maintaining high-resource language performance. We discover that effective transfer operates through functional divergence rather than strict alignment; performance gains correlate with increased language cluster separation. Our results demonstrate that targeted activation steering can unlock latent multilingual capacity in existing models without modification to model weights.

An increasing body of work has leveraged multilingual language models for Natural Language Generation tasks such as summarization. A major empirical bottleneck in this area is the shortage of accurate and robust evaluation metrics for many languages, which hinders progress. Recent studies suggest that multilingual language models often use English as an internal pivot language, and that misalignment with this pivot can lead to degraded downstream performance. Motivated by the hypothesis that this mismatch could also apply to multilingual neural metrics, we ask whether steering their activations toward an English pivot can improve correlation with human judgments. We experiment with encoder- and decoder-based metrics and find that test-time intervention methods are effective across the board, increasing metric effectiveness for diverse languages.

Controllable generation requires language models to realize output characteristics such as reading level, politeness, and toxicity. Existing steering methods are often indirect, require access to internal activations, or depend on auxiliary trained models. We propose SWAI, a training-free inference-time method that addresses these limitations by steering directly in logit space using corpus-derived token statistics. SWAI computes z-normalized one-vs-rest log-odds scores from labeled corpora and biases high-scoring tokens only within the model's top-K candidate set, allowing control to favor target-characteristic tokens while preserving contextually plausible choices. Across readability, politeness, and toxicity control, SWAI consistently improves over prompt-based and prior logit-level baselines without modifying model parameters, accessing internal layers, or training an auxiliary model. Selectivity and lookup-table ablations show that the gains come from target-specific statistical scores rather than generic logit perturbation. These results indicate that effective steering does not require learned controllers when the logit intervention is guided by target-specific statistics under high-probability candidates.

Controlling emergent behavioral personas (e.g., sycophancy, hallucination) in Large Language Models (LLMs) is critical for AI safety, yet remains a persistent challenge. Existing solutions face a dilemma: manual prompt engineering is intuitive but unscalable and imprecise, while automatic optimization methods are effective but operate as "black boxes" with no interpretable connection to model internals. We propose a novel framework that adapts gradient ascent to LLMs, enabling targeted prompt discovery. In specific, we propose two methods, RESGA and SAEGA, that both optimize randomly initialized prompts to achieve better aligned representation with an identified persona direction. We introduce fluent gradient ascent to control the fluency of discovered persona steering prompts. We demonstrate RESGA and SAEGA's effectiveness across Llama 3.1, Qwen 2.5, and Gemma 3 for steering three different personas, sycophancy, hallucination, and myopic reward. Crucially, on sycophancy, our automatically discovered prompts achieve significant improvement (49.90% compared with 79.24%). By grounding prompt discovery in mechanistically meaningful features, our method offers a new paradigm for controllable and interpretable behavior modification. We release our scripts for RESGA and SAEGA in this github repo: https://github.com/HarshSaini10/RESGA_SAEGA.

Multilingual Large Language Models (LLMs) often exhibit hallucinations such as unintended code-switching, reducing reliability in downstream tasks. We propose latent-space language steering, a lightweight inference-time method that identifies language directions via Principal Component Analysis (PCA) on parallel translations and steers token embeddings along these axes to control language identity. Our approach mitigates code-switching (reduced next-token distributional divergence by up to 55%, reduced Code-Switching Index by 63--99%) while preserving broad semantic direction with negligible computational overhead and requires only minimal parallel data for calibration. Additionally, we evaluate language classification using a single principal component (96-100% accuracy) and further analyze the layer-wise evolution of language representations.

The ability to control LLMs' emulated emotional states and personality traits is an essential step in enabling rich, human-centered interactions in socially interactive settings. We introduce PsySET, a Psychologically-informed benchmark to evaluate LLM Steering Effectiveness and Trustworthiness across the emotion and personality domains. Our study spans four models from different LLM families paired with various steering strategies, including prompting, fine-tuning, and representation engineering. Our results indicate that prompting is consistently effective but limited in intensity control, whereas vector injections achieve finer controllability while slightly reducing output quality. Moreover, we explore the trustworthiness of steered LLMs by assessing safety, truthfulness, fairness, and ethics, highlighting potential side effects and behavioral shifts. Notably, we observe idiosyncratic effects; for instance, even a positive emotion like joy can degrade robustness to adversarial factuality, lower privacy awareness, and increase preferential bias. Meanwhile, anger predictably elevates toxicity yet strengthens leakage resistance. Our framework establishes the first holistic evaluation of emotion and personality steering, offering insights into its interpretability and reliability for socially interactive applications.

Existing data poisoning attacks on retrieval-augmented generation (RAG) systems scale poorly because they require costly optimization of poisoned documents for each target phrase. We introduce Eyes-on-Me, a modular attack that decomposes an adversarial document into reusable Attention Attractors and Focus Regions. Attractors are optimized to direct attention to the Focus Region. Attackers can then insert semantic baits for the retriever or malicious instructions for the generator, adapting to new targets at near zero cost. This is achieved by steering a small subset of attention heads that we empirically identify as strongly correlated with attack success. Across 18 end-to-end RAG settings (3 datasets $\times$ 2 retrievers $\times$ 3 generators), Eyes-on-Me raises average attack success rates from 21.9 to 57.8 (+35.9 points, 2.6$\times$ over prior work). A single optimized attractor transfers to unseen black box retrievers and generators without retraining. Our findings establish a scalable paradigm for RAG data poisoning and show that modular, reusable components pose a practical threat to modern AI systems. They also contribute to interpretability research by revealing a strong link between attention concentration and model outputs.

As Vision Language Models (VLMs) are deployed across safety-critical applications, understanding and controlling their behavioral patterns has become increasingly important. Existing behavioral control methods face significant limitations: system prompting approaches could easily be overridden by user instructions, while applying activation-based steering vectors requires invasive runtime access to model internals, precluding deployment with API-based services and closed-source models. Finding steering methods that transfer across multiple VLMs is still an open area of research. To this end, we introduce universal visual input based steering for output redirection (VISOR++), to achieve behavioral control through optimized visual inputs alone. We demonstrate that a single VISOR++ image can be generated for an ensemble of VLMs to emulate each of their steering vectors. By crafting universal visual inputs that induce target activation patterns, VISOR++ eliminates the need for runtime model access while remaining deployment-agnostic. This means that when an underlying model supports multimodal capability, model behaviors can be steered by inserting an image input replacing runtime steering vector based interventions. We first demonstrate the effectiveness of the VISOR++ images on open-access models such as LLaVA-1.5-7B and IDEFICS2-8B along three alignment directions: refusal, sycophancy and survival instinct. Both the model-specific steering images and the jointly optimized images achieve performance parity closely following that of steering vectors for both positive and negative steering tasks. We also show the promise of VISOR++ images in achieving directional behavioral shifts for unseen models including both open-access and closed-access ones. Furthermore, VISOR++ images are able to preserve 99.9% performance on 14,000 unrelated MMLU evaluation tasks.

Large language model (LLM) steering has emerged as a promising paradigm for controlling model behavior at inference time through targeted manipulation of hidden states, offering a lightweight alternative to expensive retraining. However, existing steering frameworks suffer from critical limitations: computational inefficiency, limited extensibility, and restricted functionality that hinder both research progress and practical deployment. We present EasySteer, a unified framework for high-performance, extensible LLM steering built on vLLM. Our system features modular architecture with pluggable interfaces for both analysis-based and learning-based methods, fine-grained parameter control, pre-computed steering vectors for eight application domains, and an interactive demonstration system. Through deep integration with vLLM's optimized inference engine, EasySteer achieves 10.8-22.3$\times$ speedup over existing frameworks. Extensive experiments demonstrate its effectiveness in overthinking mitigation, hallucination reduction, and other key applications. EasySteer transforms steering from research technique to production-ready capability, establishing critical infrastructure for deployable, controllable language models.

Large language models (LLMs) increasingly serve as automated evaluators, yet they suffer from "self-preference bias": a tendency to favor their own outputs over those of other models. This bias undermines fairness and reliability in evaluation pipelines, particularly for tasks like preference tuning and model routing. We investigate whether lightweight steering vectors can mitigate this problem at inference time without retraining. We introduce a curated dataset that distinguishes self-preference bias into justified examples of self-preference and unjustified examples of self-preference, and we construct steering vectors using two methods: Contrastive Activation Addition (CAA) and an optimization-based approach. Our results show that steering vectors can reduce unjustified self-preference bias by up to 97%, substantially outperforming prompting and direct preference optimization baselines. Yet steering vectors are unstable on legitimate self-preference and unbiased agreement, implying self-preference spans multiple or nonlinear directions. This underscores both their promise and limits as safeguards for LLM-as-judges and motivates more robust interventions.

Reasoning LLMs trained with long chain-of-thought often overthink: they spend tokens on redundant reflection and transitions that inflate cost without improving accuracy. Static activation steering (e.g.\ SEAL) suppresses such content with a fixed vector, but applies the same strength regardless of how redundant the current chunk actually is. We describe PID-steering, a training-free, decoding-time method that modulates the steering strength with a PID controller driven by a lightweight chunk-level redundancy classifier. On a subset of GSM8K with DeepSeek-R1-Distill-Qwen-1.5B, the method improves accuracy from 85.7% to 89.6% (+3.9 pp) while cutting average output length from 1026 to 790 tokens ($-$23%). We report it as a small-scale proof of concept rather than a benchmark result.

Inference-time steering aims to alter a large language model's (LLM's) responses without changing its parameters, but a central challenge is identifying the internal modules that most strongly govern the target behavior. Existing approaches often rely on simplistic cues or ad hoc heuristics, leading to suboptimal or unintended effects. We introduce REAL, a framework for identifying behavior-relevant modules (attention heads or layers) in Transformer models. For each module, REAL trains a vector-quantized autoencoder (VQ-AE) on its hidden activations and uses a shared, learnable codebook to partition the latent space into behavior-relevant and behavior-irrelevant subspaces. REAL quantifies a module's behavioral relevance by how well its VQ-AE encodings discriminate behavior-aligned from behavior-violating responses via a binary classification metric; this score guides both module selection and steering strength. We evaluate REAL across eight LLMs from the Llama and Qwen families and nine datasets spanning truthfulness enhancement, open-domain QA under knowledge conflicts, and general alignment tasks. REAL enables more effective inference-time interventions, achieving an average relative improvement of 20% (up to 81.5%) over the ITI method on truthfulness steering. In addition, the modules selected by REAL exhibit strong zero-shot generalization in cross-domain truthfulness-steering scenarios.

The prevalence of Large Language Models (LLMs) in critical applications highlights the need for controlled language generation methods that are both computationally efficient and enjoy performance guarantees. To address this need, we use a common model of concept semantics as linearly represented in an LLM's latent space. In particular, we take the view that natural language generation traces a trajectory in this continuous semantic space, realized by the language model's hidden activations. This view permits a control-theoretic treatment of text generation in latent space, in which we propose Linear Semantic Control (LiSeCo), a lightweight, gradient-free intervention that dynamically steers trajectories away from regions corresponding to undesired meanings. In particular, we propose to directly intervene, in an online fashion, the activations of the token that is being generated in embedding space. Crucially, LiSeCo does not simply steer activations towards a desirable region. Instead, it relies on classical techniques from control theory to precisely control activations in a context-dependent way, and guarantees that they are brought into a specific pre-defined region of embedding space that corresponds to allowed semantics. The intervention is computed in closed form according to an optimal controller formulation, minimally impacting generation time. This control of the activations in embedding space allows for fine-grained steering of attributes of the generated sequence. We demonstrate that our approach is effective on different tasks -- toxicity, sentiment, and language (English/Spanish) steering -- while maintaining text quality.