2026-09-02 04:00 UTCOriginal source2 min readUpdated: 2026-09-02 09:07 UTC

CoLT-Drive: Counterfactual Long-Tail Benchmarking and Knowledge-Preserving Adaptation for Driving Affordance Prediction

arXiv:2609.00242v1 Announce Type: new Abstract: Long-tail autonomous driving failures are often framed as rare-object recognition errors. We argue that this view is incomplete: the decision-critical question is not only whether a model recognizes an unusual object, but whether it infers how that object changes the ego vehicle's feasible high-level actions. We formalize this problem as decision-level driving affordance prediction, where a model maps a front-view image, ego-motion history, and navigation command to a structured longitudinal--lateral meta-action. To evaluate this capability, we introduce CoLT-Drive, a 3,536-sample counterfactual long-tail benchmark that inserts rare objects into otherwise fixed driving scenes and measures whether models predict acceptable action pairs. To improve deployable small VLMs, we propose KPA, a knowledge-preserving adaptation framework that combines structured perception-to-decision prompting, SLERP-based expert merging, and RegMoE, a regime-aware LoRA mixture-of-experts module. KPA preserves the pretrained model's open-world knowledge while allocating lightweight adaptation capacity to different driving decision regimes. Experiments on an in-domain driving split and CoLT-Drive show that KPA achieves 60.8\% pair accuracy on CoLT-Drive, outperforming the pretrained Qwen3-VL-2B baseline (50.3\%) and LoRA SFT (32.4\%) while maintaining competitive in-domain accuracy. Our benchmark and code are available at https://huggingface.co/datasets/tangzx2024/CoLT-Drive and https://github.com/tangzhengxu/CoLT-Drive.

SourcearXiv Computer VisionAuthor: Zhengxu Tang, Guofeng Cui, Ziyu Gong, Xiaozhou Zhang, Ruifeng Deng, Chengzhi Qi, Ke Chen, Sachin Patil, Tianjun Xiao, Langechuan Liu, Pichao Wang

-->

[Submitted on 31 Aug 2026]

Title:CoLT-Drive: Counterfactual Long-Tail Benchmarking and Knowledge-Preserving Adaptation for Driving Affordance Prediction

View a PDF of the paper titled CoLT-Drive: Counterfactual Long-Tail Benchmarking and Knowledge-Preserving Adaptation for Driving Affordance Prediction, by Zhengxu Tang and Guofeng Cui and Ziyu Gong and Xiaozhou Zhang and Ruifeng Deng and Chengzhi Qi and Ke Chen and Sachin Patil and Tianjun Xiao and Langechuan Liu and Pichao Wang

View PDF HTML (experimental)

Abstract:Long-tail autonomous driving failures are often framed as rare-object recognition errors. We argue that this view is incomplete: the decision-critical question is not only whether a model recognizes an unusual object, but whether it infers how that object changes the ego vehicle's feasible high-level actions. We formalize this problem as decision-level driving affordance prediction, where a model maps a front-view image, ego-motion history, and navigation command to a structured longitudinal--lateral meta-action. To evaluate this capability, we introduce CoLT-Drive, a 3,536-sample counterfactual long-tail benchmark that inserts rare objects into otherwise fixed driving scenes and measures whether models predict acceptable action pairs. To improve deployable small VLMs, we propose KPA, a knowledge-preserving adaptation framework that combines structured perception-to-decision prompting, SLERP-based expert merging, and RegMoE, a regime-aware LoRA mixture-of-experts module. KPA preserves the pretrained model's open-world knowledge while allocating lightweight adaptation capacity to different driving decision regimes. Experiments on an in-domain driving split and CoLT-Drive show that KPA achieves 60.8\% pair accuracy on CoLT-Drive, outperforming the pretrained Qwen3-VL-2B baseline (50.3\%) and LoRA SFT (32.4\%) while maintaining competitive in-domain accuracy. Our benchmark and code are available at this https URL and this https URL.

Comments: Accepted by EMNLP 2026

Subjects:

Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Robotics (cs.RO)

Cite as: arXiv:2609.00242 [cs.CV]

(or arXiv:2609.00242v1 [cs.CV] for this version)

https://doi.org/10.48550/arXiv.2609.00242

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Pichao Wang [view email] [v1] Mon, 31 Aug 2026 18:44:30 UTC (11,237 KB)

Full-text links:

Access Paper:

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CV

new | recent | 2026-09

Change to browse by:

cs cs.AI cs.CL cs.RO

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)