Papers with training-free approach
Shallow Focus, Deep Fixes: Enhancing Shallow Layers Vision Attention Sinks to Alleviate Hallucination in LVLMs (2025.emnlp-main)
Copied to clipboard
Xiaofeng Zhang, Yihao Quan, Chen Shen, Chaochen Gu, Xiaosong Yuan, Shaotian Yan, Jiawei Cao, Hao Cheng, Kaijie Wu, Jieping Ye
| Challenge: | Multimodal large language models (MLLMs) demonstrate excellent abilities for understanding visual information, but the hallucination remains a challenging problem. |
| Approach: | They propose a training-free approach to enhance vision attention sinks to facilitate convergence of the image token attention sink within shallow layers. |
| Outcome: | The proposed approach improves the convergence of the image token attention sink within shallow layers and strengthens the layer’s focus on the image itself. |
Cracking the Code of Hallucination in LVLMs with Vision-aware Head Divergence (2025.acl-long)
Copied to clipboard
Jinghan He, Kuan Zhu, Haiyun Guo, Junfeng Fang, Zhenglin Hua, Yuheng Jia, Ming Tang, Tat-Seng Chua, Jinqiao Wang
| Challenge: | Existing methods focus on alignment training or decoding refinements but address symptoms at the generation stage without probing the underlying causes. |
| Approach: | They propose a training-free approach to mitigate hallucination by enhancing the role of vision-aware attention heads. |
| Outcome: | The proposed method achieves superior performance compared to state-of-the-art approaches in mitigating hallucinations while maintaining high efficiency with negligible additional time overhead. |
VReST: Enhancing Reasoning in Large Vision-Language Models through Tree Search and Self-Reward Mechanism (2025.acl-long)
Copied to clipboard
| Challenge: | Large Vision-Language Models (LVLMs) have shown exceptional performance in multimodal tasks, but their effectiveness in complex visual reasoning is constrained. |
| Approach: | They propose a training-free approach that enhances Reasoning in Large Vision-Language Models . they propose integrating Monte Carlo Tree Search and Self-Reward mechanisms into the reasoning tree . |
| Outcome: | The proposed approach surpasses current prompting methods and secures state-of-the-art performance across three multimodal reasoning benchmarks. |
QAEncoder: Towards Aligned Representation Learning in Question Answering Systems (2025.acl-long)
Copied to clipboard
Zhengren Wang, Qinhan Yu, Shida Wei, Zhiyu Li, Feiyu Xiong, Xiaoxing Wang, Simin Niu, Hao Liang, Wentao Zhang
| Challenge: | Modern QA systems entail retrieval-augmented generation (RAG) for accurate and trustworthy responses, but the inherent gap between user queries and relevant documents hinders precise matching. |
| Approach: | They propose a retrieval-augmented generation (RAG)-based approach to bridge this gap by attaching document fingerprints to the embedding to estimate the expectation of potential queries. |
| Outcome: | Experiments across diverse datasets, languages, and embedding models confirm the proposed solution is simple-yet-effective with zero additional index storage, retrieval latency, training costs, or catastrophic forgetting and hallucination issues. |
GASE: Generatively Augmented Sentence Encoding (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Generatively Augmented Sentence Encoding variates the input text by paraphrasing, summarizing, or extracting keywords, followed by pooling the original and synthetic embeddings. |
| Approach: | They propose a training-free approach to improve sentence embeddings by applying generative text models for data augmentation at inference time. |
| Outcome: | The proposed approach does not require access to model parameters or computational resources typically required for fine-tuning state-of-the-art models. |
Generative Explore-Exploit: Training-free Optimization of Generative Recommender Systems using LLM Optimizers (2024.acl-long)
Copied to clipboard
Lütfi Kerem Senel, Besnik Fetahu, Davis Yoshida, Zhiyu Chen, Giuseppe Castellucci, Nikhita Vedula, Jason Ingyu Choi, Shervin Malmasi
| Challenge: | Large Language Models (LLMs) have given rise to generative recommenders . however, improving the generated content through user feedback is prohibitively expensive . |
| Approach: | They propose a generative explore-exploit method that exploits items with high engagement and actively explores hidden population preferences to improve recommendation quality. |
| Outcome: | The proposed approach exploits items with high engagement and actively explores hidden population preferences to improve recommendation quality. |
EditID: Training-Free Editable ID Customization for Text-to-Image Generation (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing text-to-image models for customized IDs focus on ID consistency while neglecting editability. |
| Approach: | They propose a training-free approach to editable customized IDs based on the DiT architecture . EditID deconstructs existing text-to-image models into image generation branch and character feature branch . |
| Outcome: | The proposed solution achieves high-quality images with editable IDs while maintaining ID consistency. |
RASPberry: Retrieval-Augmented Monte Carlo Tree Self-Play with Reasoning Consistency for Multi-Hop Question Answering (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for generating and analyzing multiple document knowledge are not effective for multi-hop question answering. |
| Approach: | They propose a Monte Carlo tree-based approach to inference-time scaling using RASPberry. |
| Outcome: | Experimental results show that the proposed method achieves better inference-time scaling on smaller LLMs. |
Activation Steering Decoding: Mitigating Hallucination in Large Vision-Language Models through Bidirectional Hidden State Intervention (2025.acl-long)
Copied to clipboard
| Challenge: | Large Vision Language Models (LVLMs) suffer from hallucination where generated textual descriptions fail to align accurately with visual semantics. |
| Approach: | They propose a training-free approach that mitigates hallucination through targeted intervention in the model’s intermediate activations by identifying directional patterns of hallucinism in the activation space using a small calibration set. |
| Outcome: | The proposed approach reduces hallucination across multiple benchmarks while maintaining performance on general visual understanding tasks. |
START: Self-taught Reasoner with Tools (2025.emnlp-main)
Copied to clipboard
Chengpeng Li, Mingfeng Xue, Zhenru Zhang, Jiaxi Yang, Beichen Zhang, Bowen Yu, Binyuan Hui, Junyang Lin, Xiang Wang, Dayiheng Liu
| Challenge: | Large Reasoning Models (LRMs) have demonstrated remarkable capabilities in complex reasoning through long chain-of-thought, yet they struggle with precise computations and algorithmic operations. |
| Approach: | They propose a training-free approach that activates LRMs’ latent tool-use capabilities through artificial hints and a framework that enables models to learn effective tool utilization through diverse hint patterns and rejection-based data synthesis. |
| Outcome: | Experiments show that START significantly improves state-of-the-art LRMs across challenging benchmarks, including competition-level mathematics (AMC23: 95.0%, AIME24: 75.6%) and graduate-level science questions (GPQA: 64.6%). |
Hire a Linguist!: Learning Endangered Languages in LLMs with In-Context Linguistic Descriptions (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing LLMs rarely perform well in unseen, endangered languages . Existing models such as Llama and GPT-4 lack a rich corpus of training data . |
| Approach: | They propose a training-free approach to enable an LLM to process unseen languages that hardly occur in its pre-training. |
| Outcome: | The proposed approach elevates translation capability from GPT-4’s 0 to 10.5 BLEU for 10 language directions. |
TIDES: Technical Information Discovery and Extraction System (2025.emnlp-main)
Copied to clipboard
| Challenge: | Traditional pre-trained LLMs struggle with domain-specific terminology, while fine-tuned LLM requires substantial computational resources. |
| Approach: | They propose a training-free approach that combines TF-IDF with prompt-based LLMs to address technical questions. |
| Outcome: | The proposed system improves the accuracy and efficiency of QA systems in technical domains without LLM retraining. |
Decoding on Graphs: Faithful and Sound Reasoning on Knowledge Graphs through Generation of Well-Formed Chains (2025.acl-long)
Copied to clipboard
| Challenge: | Existing research on the utilization of Knowledge Graphs (KGs) for large language models (LLMs) relies on subgraph retriever or iterative prompting, overlooking the potential synergy of LLMs’ step-wise reasoning capabilities and KGs’ structural nature. |
| Approach: | They propose a graph-aware constrained decoding framework that facilitates a deep synergy between LLMs and KGs by constraint derived from the topology of the KG. |
| Outcome: | The proposed framework can provide faithful and sound reasoning for KGQA. |
Multi-Modality Expansion and Retention for LLMs through Parameter Merging and Decoupling (2025.acl-long)
Copied to clipboard
Junlin Li, Guodong Du, Jing Li, Sim Kuan Goh, Wenya Wang, Yequan Wang, Fangming Liu, Ho-Kin Tang, Saleh Alharbi, Daojing He, Min Zhang
| Challenge: | Large Language Models (LLMs) are a cornerstone in artificial intelligence due to their exceptional performance. |
| Approach: | They propose a training-free approach that integrates existing MLLMs for effective multimodal expansion while retaining their original performance. |
| Outcome: | The proposed approach can expand LLMs' multimodal capabilities while retaining original performance. |
Discovering a Shared Logical Subspace: Steering LLM Logical Reasoning via Alignment of Natural-Language and Symbolic Views (2026.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to multistep logical reasoning are limited by natural language refinement or external symbolic solvers. |
| Approach: | They propose a logical subspace that captures logical reasoning capabilities in LLMs that are shared across views while remaining independent of surface forms. |
| Outcome: | The proposed approach improves accuracy by 11 percentage points and generalizes well on out-of-domain problems. |