MIRA: Mutual Information guided calibration for Reliable Test-Time Adaptation
MIRA is a source-free online adaptation method using lightweight Gaussian perturbations and mutual information to improve test-time adaptation under distribution shift.
Stories per day · 30 days
3 RSS
Relatedbenchmarks1calibration1distribution-shift1embodied-ai1exploration1imagenet1latent-optimization1llm-agents1meta-reinforcement-learning1meta-world1model-deployment1mutual-information1neural-memory1open-set1pointmaze1representation-learning1robotics1rule-drift1
10 stories, newest first · one card per story
Stories tagged #test-time-adaptation
PORTAL chooses when a test-time model intervention is worth using
The controller selects which batches to adapt, which inputs to label, and when querying should stop.
TALOR rectifies open-set test-time features without backprop
On ImageNet-C with six csOOD datasets it beats UniEnt by 2.2% H-score while using 12% of the GPU memory.
METTA adapts meta-RL agents with two-scale neural memory
On held-out Meta-World ML10/45 environments, METTA reports a success rate over 50% at a low training budget.
AdaExploreBench tests LLM agents on exploration and rule drift
Across eight leading models, GPT-6 Sol reaches only 61.9% calibrated accuracy when rules stay hidden and can change.
T2-z: Test-Time Optimization of Environment Latents for Robotic World Models
T2-Z adapts frozen robotic world models at test time by optimizing a compact environment latent.
Test-Time Agent Evolution for Long-Horizon Legal Reasoning
A training-free legal agent adapts at test time using memory evolution and rubric-aligned collaboration.
WISE-ATTA: When to Ask for Labels in Budgeted Active Test-Time Adaptation
WISE-ATTA schedules scarce labels during test-time adaptation and matches prior methods with fewer annotations.
Harness Learning Enables Generalizable Test-Time Adaptation
Harness learning trains a proposer to revise an agent's executable harness from execution feedback.
Minimally Invasive Steering of Language Models
MISVO steers frozen language models toward a test-time reward while limiting distribution shift via local KL geometry.