ZeroTuning: Unlocking the Initial Token's Power to Enhance Large Language Models Without Training

• URL: http://arxiv.org/abs/2505.11739v2
• Date: Fri, 26 Sep 2025 03:55:57 GMT
• Title: ZeroTuning: Unlocking the Initial Token's Power to Enhance Large Language Models Without Training
• Authors: Feijiang Han, Xiaodong Yu, Jianheng Tang, Delip Rao, Weihua Du, Lyle Ungar,
• Abstract summary: We introduce ZeroTuning: a training-free method that improves LLM performance by applying head-specific attention adjustments to the initial token.<n>We show theoretically that adding lightweight biases to this token's attention logits monotonically controls the entropy of the downstream attention distribution.<n>We present two variants: a supervised mode that calibrates on validation examples, and a novel unsupervised mode that directly minimizes the model's output entropy.
• Score: 15.783265191574392
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Token-level attention tuning, a class of training-free methods including Post-hoc Attention Steering (PASTA) and Attention Calibration (ACT), has emerged as a promising way to improve frozen LLMs with interpretable interventions. However, these methods depend on auxiliary heuristics to identify "important" task-specific tokens, which can introduce bias and limit applicability when token importance is unclear or when using optimized kernels where attention maps are inaccessible. We propose a simpler and more elegant alternative: acting only on the initial token (e.g., <BOS> in LLaMA). We show theoretically that adding lightweight biases to this token's attention logits monotonically controls the entropy of the downstream attention distribution - an effect amplified by its natural function as an attention sink. Our empirical analysis reveals that this tuning process can positively affect LLMs and better unlock their pretrained knowledge, with stronger effects in early layers and distinct scaling preferences across attention heads. Building on these insights, we introduce ZeroTuning: a training-free method that improves LLM performance by applying head-specific attention adjustments to the initial token, requiring zero parameter updates. We present two variants: a supervised mode that calibrates on validation examples, and a novel unsupervised mode that directly minimizes the model's output entropy. The method is lightweight, kernel-agnostic, and requires only four lines of modification to the standard LlamaAttention code. It achieves broad gains across 15 datasets and outperforms previous, more complex methods; for instance, with Llama-3.1-8B, it yields relative improvements of 19.9% on classification, 4.5% on question answering, and 2.1% on dialogue. ZeroTuning also works out-of-the-box with quantized inference and maintains its performance improvements with increasing context lengths.

Related papers

• Decomposing and Composing: Towards Efficient Vision-Language Continual Learning via Rank-1 Expert Pool in a Single LoRA [50.97792275353563]
  We introduce a novel framework that restructures a single Low-Rank Adaptation (LoRA) module as a decomposable Rank-1 Expert Pool.<n>Our method learns to dynamically compose a sparse, task-specific update by selecting from this expert pool, guided by the semantics of the [Guided] token.
  arXiv  Detail & Related papers  (2026-01-30T10:54:51Z)
• All You Need is One: Capsule Prompt Tuning with a Single Vector [86.68105855537762]
  Current prompt-based learning methods rely on laborious grid searching for optimal prompt length and typically require considerable number of prompts.<n>We introduce Capsule Prompt-Tuning (CaPT), an efficient and effective solution that leverages off-the-shelf, informative instance semantics into prompt-based learning.<n>Our approach innovatively integrates both instance-aware and task-aware information in a nearly parameter-free manner.
  arXiv  Detail & Related papers  (2025-10-19T00:02:59Z)
• Attention Illuminates LLM Reasoning: The Preplan-and-Anchor Rhythm Enables Fine-Grained Policy Optimization [56.083511902353365]
  Reinforcement learning (RL) typically applies uniform credit across an entire generation of Large language models.<n>This work positions attention as a privileged substrate that renders the internal logic of LLMs as a mechanistic blueprint of reasoning itself.<n>We introduce three novel RL strategies that dynamically perform targeted credit assignment to critical nodes.
  arXiv  Detail & Related papers  (2025-10-15T13:49:51Z)
• Informed Routing in LLMs: Smarter Token-Level Computation for Faster Inference [7.690958366125321]
  This paper introduces informed routing, a new paradigm that proactively addresses these issues.<n>We propose the Lightweight Feature Forecaster (LFF), a small predictive module that estimates a unit's output before routing decisions are made.<n>Experiments on both language modeling and reasoning tasks show that informed routing achieves state-of-the-art efficiency-performance trade-offs.
  arXiv  Detail & Related papers  (2025-10-10T09:59:36Z)
• Training-Free Token Pruning via Zeroth-Order Gradient Estimation in Vision-Language Models [16.540220733551823]
  Large Vision-Language Models (VLMs) enable strong multimodal reasoning but incur heavy inference costs from redundant visual tokens.<n> Attention-based methods rely on raw attention scores, which are often unstable across layers and heads.<n>We propose ours, a training-free framework built on a simple intuition.
  arXiv  Detail & Related papers  (2025-09-29T14:20:05Z)
• GrAInS: Gradient-based Attribution for Inference-Time Steering of LLMs and VLMs [56.93583799109029]
  GrAInS is an inference-time steering approach that operates across both language-only and vision-language models and tasks.<n>During inference, GrAInS hidden activations at transformer layers guided by token-level attribution signals, and normalizes activations to preserve representational scale.<n>It consistently outperforms both fine-tuning and existing steering baselines.
  arXiv  Detail & Related papers  (2025-07-24T02:34:13Z)
• IGD: Token Decisiveness Modeling via Information Gain in LLMs for Personalized Recommendation [70.2753541780788]
  We introduce an Information Gain-based Decisiveness-aware Token handling (IGD) strategy that integrates token decisiveness into both tuning and decoding.<n>IGD consistently improves recommendation accuracy, achieving significant gains on widely used ranking metrics compared to strong baselines.
  arXiv  Detail & Related papers  (2025-06-16T08:28:19Z)
• ALPS: Attention Localization and Pruning Strategy for Efficient Alignment of Large Language Models [14.657194214702473]
  We propose an efficient algorithm that localizes the most task-sensitive attention heads and prunes them by restricting attention training updates to these heads.<n> Experimental results demonstrate that our method activates only 10% of attention parameters during fine-tuning while achieving a 2% performance improvement over baselines on three tasks.
  arXiv  Detail & Related papers  (2025-05-24T17:19:34Z)
• Attention Reallocation: Towards Zero-cost and Controllable Hallucination Mitigation of MLLMs [62.9348974370985]
  We propose attention reallocation (AttnReal) to mitigate hallucinations with nearly zero extra cost.<n>Our approach is motivated by the key observations that, MLLM's unreasonable attention distribution causes features to be dominated by historical output tokens.<n>Based on the observations, AttnReal recycles excessive attention from output tokens and reallocates it to visual tokens, which reduces MLLM's reliance on language priors.
  arXiv  Detail & Related papers  (2025-03-11T11:52:37Z)
• The First Few Tokens Are All You Need: An Efficient and Effective Unsupervised Prefix Fine-Tuning Method for Reasoning Models [69.798277882245]
  We introduce Unsupervised Prefix Fine-Tuning (UPFT) to enhance large language models' reasoning efficiency.<n>UPFT removes the need for labeled data or exhaustive sampling.<n> Experiments show that UPFT matches the performance of supervised methods.
  arXiv  Detail & Related papers  (2025-03-04T18:56:03Z)
• RSQ: Learning from Important Tokens Leads to Better Quantized LLMs [65.5558181902098]
  Layer-wise quantization is a key technique for efficiently compressing large models without expensive retraining.<n>We propose RSQ (Rotate, Scale, then Quantize), which applies rotations to the model to mitigate outliers.<n>We demonstrate that RSQ consistently outperforms baseline methods across multiple downstream tasks and three model families.
  arXiv  Detail & Related papers  (2025-03-03T18:46:33Z)
• Tactic: Adaptive Sparse Attention with Clustering and Distribution Fitting for Long-Context LLMs [10.52833484759311]
  We propose Tactic, a sparsity-adaptive and calibration-free sparse attention mechanism.<n>It dynamically selects tokens based on their cumulative attention scores rather than a fixed token budget.<n>We show that Tactic outperforms existing sparse attention algorithms, achieving superior accuracy and up to 7.29x decode attention speedup.
  arXiv  Detail & Related papers  (2025-02-17T08:39:43Z)
• AttentionPredictor: Temporal Pattern Matters for Efficient LLM Inference [51.1972443343829]
  We propose AttentionPredictor, which is the first learning-based critical token identification approach.<n> AttentionPredictor accurately predicts the attention score while consuming negligible memory.<n>We also propose a cross-token critical cache prefetching framework that hides the token time overhead to accelerate the decoding stage.
  arXiv  Detail & Related papers  (2025-02-06T13:41:46Z)
• Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability [53.51560766150442]
  Critical tokens are elements within reasoning trajectories that significantly influence incorrect outcomes.<n>We present a novel framework for identifying these tokens through rollout sampling.<n>We show that identifying and replacing critical tokens significantly improves model accuracy.
  arXiv  Detail & Related papers  (2024-11-29T18:58:22Z)
• SeerAttention: Learning Intrinsic Sparse Attention in Your LLMs [10.702409298302547]
  SeerAttention learns the block-level attention sparsity from the Large Language Models itself.<n>Inspired by the gating mechanism in Mixture of Experts (MoE), SeerAttention augments the conventional attention with a learnable gate.<n>Our evaluation results demonstrate that SeerAttention achieves better model accuracy and lower latency for long-context pre-filling.
  arXiv  Detail & Related papers  (2024-10-17T07:07:09Z)
• Fine-Tuning on Diverse Reasoning Chains Drives Within-Inference CoT Refinement in LLMs [63.36637269634553]
  We introduce a novel approach where LLMs are fine-tuned to generate a sequence of Diverse Chains of Thought (DCoT) within a single inference step.<n>We show that fine-tuning on DCoT improves performance over the CoT baseline across model families and scales.<n>Our work is also significant because both quantitative analyses and manual evaluations reveal the observed gains stem from the models' ability to refine an initial reasoning chain.
  arXiv  Detail & Related papers  (2024-07-03T15:01:18Z)
• Unveiling and Harnessing Hidden Attention Sinks: Enhancing Large Language Models without Training through Attention Calibration [15.36841874118801]
  We aim to provide a more profound understanding of the existence of attention sinks within large language models (LLMs) We propose a training-free Attention Technique (ACT) that automatically optimize the attention distributions on the fly during inference in an input-adaptive manner. ACT achieves an average improvement of up to 7.30% in accuracy across different datasets when applied to Llama-30B.
  arXiv  Detail & Related papers  (2024-06-22T07:00:43Z)
• Fortify the Shortest Stave in Attention: Enhancing Context Awareness of Large Language Models for Effective Tool Use [74.72150542395487]
  An inherent waveform pattern in the attention allocation of large language models (LLMs) significantly affects their performance in tasks demanding a high degree of context awareness. To address this issue, we propose a novel inference method named Attention Buckets.
  arXiv  Detail & Related papers  (2023-12-07T17:24:51Z)
• Zero-Shot Sharpness-Aware Quantization for Pre-trained Language Models [88.80146574509195]
  Quantization is a promising approach for reducing memory overhead and accelerating inference. We propose a novel-aware quantization (ZSAQ) framework for the zero-shot quantization of various PLMs.
  arXiv  Detail & Related papers  (2023-10-20T07:09:56Z)
• Prompting classes: Exploring the Power of Prompt Class Learning in Weakly Supervised Semantic Segmentation [15.467510304266883]
  We study the impact of prompt tuning on weakly supervised semantic segmentation. We introduce a novel approach based on a PrOmpt cLass lEarning (POLE) strategy. We demonstrate that our simple, yet efficient approach achieves SOTA performance in a well-known WSSS benchmark.
  arXiv  Detail & Related papers  (2023-06-30T19:25:18Z)

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.