ICML 2025PastRobustness
Second Workshop on Test-Time Adaptation: Putting Updates to the Test! at ICML 2025
PUT at ICML 2025
- Submission deadline
- May 24, 2025, 12:00 UTCimported from OpenReview — check the website for extensions
- Submission portal
- OpenReview
- Notes
- Topics were auto-suggested and may be imprecise — edits welcome.
Accepted papers (55)
Fetched from OpenReview (v2) on 2026-06-10.
Accurate Parameter-Efficient Test-Time Adaptation for Time Series Forecasting
Adaptive Diffusion Denoised Smoothing : Certified Robustness via Randomized Smoothing with Differentially Private Guided Denoising Diffusion
Adaptive Monocular Depth Estimation with Masked Image Consistency
AdaptMI: Adaptive Skill-based In-context Math Instructions for Small Language Models
Agentic Adversarial QA for Improving Domain-Specific LLMs
An Evidence-Based Post-Hoc Adjustment Framework for Anomaly Detection Under Data Contamination
Beyond Markovian: Reflective Exploration via Bayes-Adaptive RL for LLM Reasoning
Beyond Multiple Choice: Evaluating Steering Vectors for Adaptive Free-Form Summarization
Causal Fine-Tuning of Pre-trained Language Models for Robust Test Time Adaptation
CCC: Enhancing Video Generation via Structured MLLM Feedback
Context Tuning for In-Context Optimization
Diffusion Tree Sampling: Scalable inference‑time alignment of diffusion models
Distilling Prompts at Test-Time for Multimodal Few-Shot Learning
DPCore: Dynamic Prompt Coreset for Continual Test-Time Adaptation
e3: Learning to Explore Enables Extrapolation of Test-Time Compute for LLMs
GRIP: In-Parameter Graph Reasoning through Fine-Tuning Large Language Models
Inference-Time Alignment via Hypothesis Reweighting
JEDI: The Force of Jensen-Shannon Divergence in Disentangling Diffusion Models
Keep the Alignment, Skip the Overhead: Lightweight Instruction Alignment for Continually Trained LLMs
Language Model Personalization via Reward Factorization
Language System: A Lightweight Ranking Framework for Language Models
Learning to Self-Correct through Chain-of-Thought Verification
Leto: Modeling Multivariate Time Series with Memorizing at Test Time
LIFT: Improving Long Context Understanding of Large Language Models through Long Input Fine-Tuning
Lightweight Online Adaption for Time Series Foundation Model Forecasts
LoRA-TTT: Low-Rank Test-Time Training for Vision-Language Models
MADCAT: Combating Malware Detection Under Concept Drift with Test-Time Adaptation
Mitigating Forgetting in Low Rank Adaptation
Monitoring Risks in Test-Time Adaptation
N-Gram Induction Heads for In-Context RL: Improving Stability and Reducing Data Needs
On Distributional Robustness of In-Context Learning for Text Classification
On Training-Test (Mis)alignment in Unsupervised Combinatorial Optimization: Observation, Empirical Exploration, and Analysis
Prefix-Tuning+: Modernizing Prefix-Tuning by Decoupling the Prefix from Attention
Prune ’n Predict: Optimizing LLM Decision-making with Conformal Prediction
Reasoning as an Adaptive Defense for Safety
Rejection Sampling Based Fine Tuning Secretly Performs PPO
Replacing thinking with tool usage enables reasoning in small language models
Right Question is Already Half the Answer: Fully Unsupervisedd LLM Reasoning Incentization
Scalable Defense against In-the-wild Jailbreaking Attacks with Safety Context Retrieval
Scalable Temporal Domain Generalization via Prompting
Scaling Textual Gradients via Sampling-Based Momentum
Self-Generated In-Context Examples Improve LLM Agents for Sequential Decision-Making Tasks
Shift-Aware Test Time Adaptation and Benchmarking for Time-Series Forecasting
SteeringTTA: Guiding Diffusion Trajectories for Robust Test-Time-Adaptation
SwiTTA: Switching Domain Experts and Aggregating Contextual Features Towards Realistic Test-Time Adaptation
Temporal Sampling for Forgotten Reasoning in LLMs
Test Time Adaptation Using Adaptive Quantile Recalibration
Test-Time Adaptation for Generalizable Task Progress Estimation
Test-Time Alignment of Discrete Diffusion Models with Sequential Monte Carlo
Test-time Offline Reinforcement Learning on Goal-related Experience
The Curious Language Model: Strategic Test-Time Information Acquisition
UniTTA: Unified Benchmark and Versatile Framework Towards Realistic Test-Time Adaptation
Value Conditioned Policy Fine Tuning for Test Time Domain Adaptation
When and How Unlabeled Data Provably Improve In-Context Learning
Zero-Shot Adaptation of Behavioral Foundation Models to Unseen Dynamics