Dynamically Instance-Guided Adaptation: A Backward-free Approach for Test-Time Domain Adaptive Semantic Segmentation
Bibliographic record
Abstract
In this paper, we study the application of Test-time<br> domain adaptation in semantic segmentation (TTDA-Seg)<br> where both efficiency and effectiveness are crucial. Existing<br> methods either have low efficiency (e.g., backward optimization)<br> or ignore semantic adaptation (e.g., distribution<br> alignment). Besides, they would suffer from the accumulated<br> errors caused by unstable optimization and abnormal<br> distributions. To solve these problems, we propose a novel<br> backward-free approach for TTDA-Seg, called Dynamically<br> Instance-Guided Adaptation (DIGA). Our principle is utilizing<br> each instance to dynamically guide its own adaptation<br> in a non-parametric way, which avoids the error accumulation<br> issue and expensive optimizing cost. Specifically,<br> DIGA is composed of a distribution adaptation module<br> (DAM) and a semantic adaptation module (SAM), enabling<br> us to jointly adapt the model in two indispensable<br> aspects. DAM mixes the instance and source BN statistics to encourage the model to capture robust representation. SAM<br> combines the historical prototypes with instance-level prototypes<br> to adjust semantic predictions, which can be associated<br> with the parametric classifier to mutually benefit the<br> final results. Extensive experiments evaluated on five target<br> domains demonstrate the effectiveness and efficiency of the<br> proposed method. Our DIGA establishes new state-of-theart<br> performance in TTDA-Seg. Source code is available at:<br> https://github.com/Waybaba/DIGA.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.002 |
| Science and technology studies | 0.002 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 0.004 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".