8 articles tagged with #training-free-methods. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.
Researchers introduce PaLRS, a training-free method for aligning large language models with human preferences using lightweight steering vectors extracted from residual streams. The approach requires minimal data (100+ preference pairs) and achieves better performance than standard optimization methods like DPO with significantly lower computational costs.
Researchers discovered that language model reasoning behavior is primarily controlled by specific token patterns rather than high-level instructions, leading to the development of Mid-Think, a training-free prompting technique that achieves intermediate-budget reasoning with better accuracy-efficiency tradeoffs and improves RL training performance for models like Qwen3-8B.
Researchers introduce Decoding by Perturbation (DeP), a training-free method that reduces hallucinations in multimodal large language models by applying controlled textual perturbations during decoding. The approach addresses the core issue where language priors override visual evidence, achieving improvements across multiple benchmarks without requiring model retraining or visual manipulation.
Researchers propose MGAP, a training-free decoding method that reduces hallucinations in multimodal large language models (MLLMs) by selectively suppressing language priors while preserving semantic structure. Unlike previous approaches that blindly penalize language biases, MGAP uses geometry-aware subspace projection to distinguish between helpful and harmful language priors, achieving improved hallucination suppression without degrading model coherence.
Researchers propose Chunk-Level Guided Generation, a training-free method using off-the-shelf large language models to score intermediate reasoning steps during small-model inference for mathematical problem-solving. The approach matches or outperforms specialized reward model-based systems on benchmarks like MATH and GSM8K without requiring expensive step-level training data.
Researchers introduce TunerDiT, a training-free method for improving text-to-video generation with multiple sequential events by identifying critical steering points in diffusion transformer denoising and applying progressive prompt fusion techniques. The approach achieves state-of-the-art performance across benchmark metrics while enabling fine-tuned control over video consistency versus event separation.
AdaMerge introduces a training-free method to accelerate Vision Transformers by improving token merging through salience-aware mechanisms and adaptive layer-wise compression. The approach outperforms existing token reduction methods across all computational efficiency benchmarks, maintaining superior accuracy-to-FLOPs ratios on ImageNet-1k evaluations.
Researchers propose AnchorDiff, a training-free method for improving concept grounding in Multi-Modal Diffusion Transformers by addressing 'concept leakage' where attention activations overlap on visually similar objects. The approach uses anchor-based graph propagation to better localize and distinguish between confusable concepts, with evaluation on a newly introduced Multi-Concept Confusion Dataset.
