Prompt Tuning

Follow this topic Ask about this topic

Prompt tuning studies how small learned vectors, inserted into a model’s inputs or internal layers, can adapt a largely frozen pretrained model to new tasks. Common approaches include soft prompts and prefixes, structured or task-specific prompt banks, and routing or test-time methods that select or adjust prompts for each input or domain. These techniques are used with language and vision-language models as well as audio, graphs, and time series, often aiming to reduce training and memory costs while retaining pretrained capabilities. A central challenge is balancing effective adaptation with robustness and generalization: prompts can overfit limited or noisy data, interfere with existing behavior, or behave unpredictably across shifts, and their internal effects are not always well understood.

Latest papers

76 papers10 papers in the last four weeks, up 100% on the four weeks before. 0.1% of all new papers.

Reasoning or Just Prompting? Probing the Mechanism of Latent Chemical Reasoning
Oct 25, 2026Kaiyuan Lu, Jiatong Li, Yatao BianChemical Reasoning in LLMs Implicit Reasoning in Language Models
National University of Singapore
Robust Soft Skill Adaptation via Behavioral Localization in Tool-Using Agents
Oct 21, 2026Zijian Jiang, Xilu WangLLM Agent Skill Learning Prompt Tuning
School of Computer Science University of Surrey Guildford, Surrey GU2 7XH, UK
Not All Drift Is a Mechanism: Tracing Prompt Adaptation in LLMs
Oct 13, 2026Mikhail Aleksandrov, Nail Bashirov, Maksim Ivanov +6Prefix Tuning LLM Prompting
BRAIn Lab
SoftGene: Protein Language Model-Enhanced Soft Prompting for Interpretable Gene Set Annotation
Oct 2, 2026Drew Ross, Arya Hadizadeh Moghaddam, Dongjie Wang +2LLM-Assisted Annotation Genomics
Electrical Engineering and Computer Science, University of Kansas, USA · Cell Biology and Physiology, University of Kansas Medical Center, USA
Unified Shared Encoder in Spoof-Aware Speaker Verification with Hybrid Wavelet Prompt Tuning
Oct 1, 2026Aref Farhadipour, Srikanth Madikeri, Teodora Vukovic +2Speaker Verification ASVspoof
Department of Computational Linguistics, University of Zurich, Switzerland · Idiap Research Institute, Switzerland · Brno University of Technology, Czech Republic
UZH-CL at ArA-DF 2026: Prompt-Tuned Foundation Models and Track-Adaptive Score Fusion for Arabic Speech Deepfake Detection
Oct 1, 2026Aref Farhadipour, Teodora Vukovic, Petr MotlicekAudio Deepfake Detection Prompt Tuning
Department of Computational Linguistics, University of Zurich, Switzerland · Idiap Research Institute, Switzerland · Brno University of Technology, Czech Republic
AnchorPrompt: Self-Distilled Soft Prompts for Robust Audio-Language Models
Sep 30, 2026Pooneh Mousavi, Amir Ivry, Mirco Ravanelli +1Audio-Language Models Prompt Tuning
Concordia University · Mila – Quebec AI Institute · Technion – IIT +1
Ready2Blend: From Natural-Language Instructions to Composable Alignment Prompts
Sep 30, 2026Jeesu Jung, Hwan Chang, Juseon Do +5Continual Learning for LLMs LLM Alignment
Korea Advanced Institute of Science and Technology · Samsung SDS
When Can Prefixes Compile LoRA? Exact Resource-Capped Tests for Frozen Attention
Sep 29, 2026Joyanta Jyoti Mondal, Ibne Farabi ShihabPrefix Tuning Low-Rank Adaptation
Department of Computer and Information Sciences, University of Delaware, USA · Department of Computer Science, Iowa State University, USA
Instance-Adaptive Prompts as Context for Time-Series Foundation Models
Sep 28, 2026Zehao Xiao, Shifeng Xie, Lei Zan +5Time Series Forecasting Prompt Tuning
Huawei Noah’s Ark Lab, Paris, France · LIPADE, Universit´e Paris Cit´e, Paris, France · Huawei Noah’s Ark Lab, Shenzhen, China
Learning an Anchored Prompt Space for Continual Adaptation of Large Language Models
Sep 26, 2026Rongguang Ye, Zhan Zhuang, Yichen Wu +2Continual Learning for LLMs Prompt Tuning
City University of Hong Kong · Southern University of Science and Technology · Harvard University
Your Model Already Knows Don't Teach It, Learn to Ask It: Soft Prompting for Few-Shot Adaptation of Vision-Language Models
Sep 10, 2026Gautam Rajendrakumar Gare, Siyi Li, Hewei Wang +5Few-Shot Object Detection Few-Shot Domain Adaptation
Carnegie Mellon University · Apple
Layer-Wise Gate-Controlled Prompt Truncation in a Multimodal Chest X-Ray Classifier
Sep 6, 2026Jingtao Lei, Hongji Li, Dexiang ShuPrompt Tuning Multimodal Classification
Central South University, Changsha 410083, China · Hong Kong Baptist University, Hong Kong, China · University of Minnesota, Minneapolis, MN 55414, USA
Vision-Guided Text Prompt Tuning for Multimodal Sentiment Analysis
Sep 6, 2026Xiaoran Kou, Jingyi Wu, Peng Sun +2Multimodal Sentiment Analysis Prompt Tuning
College of Electronic and Information Engineering, Tongji University, Shanghai 201804, China · College of Intelligent Robotics and Advanced Manufacturing, Fudan University, Shanghai 200433, China · Division of Natural and Applied Sciences, Duke Kunshan University, Suzhou 215316, China
Compositional Spectral Prompts for LLM-based Online Time Series Forecasting
Sep 2, 2026Seungyoon Choi, Hyunchul Kim, Jae-Gil Lee +1Online Time Series Forecasting Time Series Forecasting
KAIST Daejeon, Republic of Korea
HiVe: Beyond Static Prompts for Multitask Learning via Hierarchy-based Vertical Mixture-of-Experts
Aug 30, 2026Hyeonjik Bae, Minyeol Kim, Susik YoonMulti-Task Learning Prompt Tuning
Computer Science and Engineering Korea University, Seoul, Korea
SPARK-SAM: Learning How to Prompt and Respond for Infrared Small Target Segmentation
Aug 21, 2026Aji Mao, Zhenming Peng, Bailin Mu +1Image Segmentation Promptable Image Segmentation
School of Information and Communication Engineering, University of Electronic Science and Technology of China
Learning What Not to Learn: Adversarial Disentangled Prompt Tuning for Robust Vision-Language Models
Aug 18, 2026Yang Chen, Zhan Zhuang, Yanbin Wei +3Prompt Tuning Adversarial Attacks on VLMs
Southern University of Science and Technology Shenzhen, China · City University of Hong Kong Hong Kong, China · Hong Kong University of Science and Technology Hong Kong, China
DSPrompt: Dynamic Soft Prompt Defense Against M-RAG Corruption
Aug 17, 2026Chang Liu, Yuni Lai, Mingyue Cui +5RAG Robustness RAG Poisoning Attacks
School of Computer Science and Technology, Xidian University · The Hong Kong Polytechnic University · Tencent Jarvis Lab
From Generalist to Specialist: A Context-Fusion Framework for Endoscopic Polyp Reporting with a Frozen VLM
Aug 16, 2026Ruijie Yang, Yan Zhu, Peiyao Fu +7VLM Adaptation Medical VLMs
Zhejiang University, Hangzhou, China · Shanghai Institute for Advanced Study, Zhejiang University, Shanghai, China · Shanghai Key Laboratory of MICCAI, Shanghai, China +4
HiRoute: Hierarchical Routed Prompt Tuning for Safety Alignment of Large Language Models
Aug 13, 2026Fangzhou Chen, Shiji Zhao, Mengyang Wang +4LLM Safety Alignment Prompt Tuning
Institute of Artificial Intelligence, Beihang University
Hate Speech Classification In Roman Urdu: A Comparative Study On Parameter Efficient Fine-Tuning And Prompt Engineering
Aug 10, 2026Toneema ZubairHate Speech Detection Prompt Engineering
Department of Computer Science Information Technology University, Lahore Pakistan
PreGress: Ranking-Native Pre-training and Prompting for Graph Node Ranking
Aug 10, 2026Lujie Ban, Jiasheng shi, Yingli Zhou +5Learning to Rank Prompt Tuning
School of Data Science, The Chinese University of Hong Kong, Shenzhen, China · GTS, Huawei Technologies Co., Ltd., China
SkillSmith: Learning to Compose Parametric Skills and Textual Knowledge
Jul 29, 2026Lucio M. Dery, Benedict Aaron Tjandra, Siavash Samiei +4Knowledge Augmentation for Language Models Prompt Tuning
Google DeepMind
Point Ladder Tuning: Parameter-Efficient Hierarchical Adaptation for 3D Point Cloud Understanding
Jul 21, 2026Junlin Chang, Longhao Zou, Rui Li3D Point Cloud Segmentation Point Cloud Classification
Beihang University, Beijing · Pengcheng Laboratory, Shenzhen
Logical Judgments Under Pressure: Diagnosing Syllogistic Stability with Learned Soft Prefixes
Jul 20, 2026Brian K ChenLogical Reasoning Symbolic Reasoning
National University of Singapore
Map as a Prompt: Learning Multi-Modal Spatial-Signal Foundation Models for Cross-scenario Wireless Localization
Jul 17, 2026Yong Chu, Xun Zhou, Zenglin Xu +2Wireless Foundation Models Prompt Tuning
Harbin Institute of Technology, Shenzhen · Pengcheng Laboratory · Shanghai Academy of AI for Science +1
Toward Federated Multimodal Graph Foundation Models: A Topology-Aware Multimodal Alignment Framework
Jul 17, 2026Xunkai Li, Guohao Fu, Yuming Ai +4Multimodal Federated Learning Cross-Modal Alignment
Beijing Institute of Technology, Beijing, 100811, China
Latent Personal Memory: Represent personal memory as dynamic soft prompts
Jun 18, 2026Debrup Das, Avinash Amballa, Yashas Malur Saidutta +3LLM Inference Efficiency Memory-Efficient Fine-Tuning
University of Massachusetts Amherst · Samsung Research America
SoftSkill: Behavioral Compression for Contextual Adaptation
Jun 18, 2026Xijia Tao, Yihua Teng, Xinyu Fu +6Prompt Tuning LLM Agent Skill Learning
The University of Hong Kong · Huawei Research · City University of Hong Kong +1

Oct 25, 2026J/K move · Enter open · S save

Reasoning or Just Prompting? Probing the Mechanism of Latent Chemical Reasoning

Kaiyuan Lu, Jiatong Li, Yatao Bian

National University of Singapore

Chemical Reasoning in LLMs Implicit Reasoning in Language Models Continuous Latent Reasoning Prompt Tuning

Open paper AskPDF arXiv

Latent reasoning performs intermediate computation in continuous space rather than through explicit tokens, which is particularly appealing in chemistry, where errors in explicit reasoning can propagate to the final answer. Yet if latent states truly constitute an implicit reasoning chain, they should be causally important. We find otherwise: swapping or moderately perturbing intermediate latent states has little effect on performance, while removing them entirely sometimes causes a substantial score drop. Thus, the latent states matter, but their molecule-specific content appears to matter far less. This points to a simpler interpretation. Rather than encoding a step-by-step reasoning chain, latent states may primarily steer subsequent model computation, much like soft prompts. Indeed, a single fixed prefix shared across molecules reproduces LatentChem's performance, and a learned static soft prompt performs comparably to per-input latent generation in controlled models. The tolerance of latent sequences to substitution and moderate perturbation further supports this connection. We therefore suggest that latent chemical reasoning may be better viewed as soft prompting than as a reasoning chain hidden in continuous space.