Copertina

When Adaptation Fails: A Gradient-Based Diagnosis of Collapsed Gating in Vision-Language Prompt Learning

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Fang, Yunxuan, Zhang, Ziwei, Wang, Xinhe
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

A Controlled Diagnostic Study of Hardware-Induced Distortions in Hardware-Aware Training
di: Fang, Yunxuan, et al.
Pubblicazione: (2026)

When Flat Minima Fail: Characterizing INT4 Quantization Collapse After FP32 Convergence
di: Armstrong, Marcus
Pubblicazione: (2026)

When Will It Fail?: Anomaly to Prompt for Forecasting Future Anomalies in Time Series
di: Park, Min-Yeong, et al.
Pubblicazione: (2025)

When Vision Fails: Text Attacks Against ViT and OCR
di: Boucher, Nicholas, et al.
Pubblicazione: (2023)

When to Stop Reusing: Dynamic Gradient Gating for Sample-Efficient RLVR
di: Miao, Yuchun, et al.
Pubblicazione: (2026)

A Tale of Two Variances: When Single-Seed Benchmarks Fail in Bayesian Deep Learning
di: Zhan, Qishi, et al.
Pubblicazione: (2026)

Gradient Starvation in Binary-Reward GRPO: Why Group-Mean Centering Fails and Why the Simplest Fix Works
di: Nie, Wenhua, et al.
Pubblicazione: (2026)

Learning to Prompt Your Domain for Vision-Language Models
di: Wei, Guoyizhe, et al.
Pubblicazione: (2023)

Overcoming the Incentive Collapse Paradox
di: Yin, Qichuan, et al.
Pubblicazione: (2026)

Signal Collapse in One-Shot Pruning: When Sparse Models Fail to Distinguish Neural Representations
di: Saikumar, Dhananjay, et al.
Pubblicazione: (2025)

Tackling Dimensional Collapse toward Comprehensive Universal Domain Adaptation
di: Fang, Hung-Chieh, et al.
Pubblicazione: (2024)

Gradient Descent Fails to Learn High-frequency Functions and Modular Arithmetic
di: Takhanov, Rustem, et al.
Pubblicazione: (2023)

When to ASK: Uncertainty-Gated Language Assistance for Reinforcement Learning
di: Monteiro, Juarez, et al.
Pubblicazione: (2026)

An Empirical Study of Federated Prompt Learning for Vision Language Model
di: Wang, Zhihao, et al.
Pubblicazione: (2025)

Improving Diversity in Language Models: When Temperature Fails, Change the Loss
di: Verine, Alexandre, et al.
Pubblicazione: (2025)

When Prompting Meets Spiking: Graph Sparse Prompting via Spiking Graph Prompt Learning
di: Jiang, Bo, et al.
Pubblicazione: (2026)

NLPrompt: Noise-Label Prompt Learning for Vision-Language Models
di: Pan, Bikang, et al.
Pubblicazione: (2024)

Gradient-Gated DPO: Stabilizing Preference Optimization in Language Models
di: Mouiche, Inoussa
Pubblicazione: (2026)

When Embedding-Based Defenses Fail: Rethinking Safety in LLM-Based Multi-Agent Systems
di: Zhang, Lingxi, et al.
Pubblicazione: (2026)

When Planning Fails Despite Correct Execution: On Epistemic Calibration for LLM-Based Multi-Agent Systems
di: Wang, Zehao, et al.
Pubblicazione: (2026)

ColA: Collaborative Adaptation with Gradient Learning
di: Diao, Enmao, et al.
Pubblicazione: (2024)

Dialogue for Prompting: a Policy-Gradient-Based Discrete Prompt Generation for Few-shot Learning
di: Li, Chengzhengxu, et al.
Pubblicazione: (2023)

Enhancing Domain Adaptation through Prompt Gradient Alignment
di: Phan, Hoang, et al.
Pubblicazione: (2024)

When Does Model Collapse Occur in Structured Interactive Learning?
di: Wu, Yuchen, et al.
Pubblicazione: (2026)

Transitive Vision-Language Prompt Learning for Domain Generalization
di: Wang, Liyuan, et al.
Pubblicazione: (2024)

When Explanations Lie: Why Many Modified BP Attributions Fail
di: Sixt, Leon, et al.
Pubblicazione: (2019)

When One Modality Rules Them All: Backdoor Modality Collapse in Multimodal Diffusion Models
di: Wang, Qitong, et al.
Pubblicazione: (2026)

Latent Domain Prompt Learning for Vision-Language Models
di: Li, Zhixing, et al.
Pubblicazione: (2025)

When Active Learning Fails, Uncalibrated Out of Distribution Uncertainty Quantification Might Be the Problem
di: Dale, Ashley S., et al.
Pubblicazione: (2025)

Noisy Test-Time Adaptation in Vision-Language Models
di: Cao, Chentao, et al.
Pubblicazione: (2025)

When Mean CE Fails: Median CE Can Better Track Language Model Quality
di: Guo, Hao, et al.
Pubblicazione: (2026)

When Models Don't Collapse: On the Consistency of Iterative MLE
di: Barzilai, Daniel, et al.
Pubblicazione: (2025)

AOI: Turning Failed Trajectories into Training Signals for Autonomous Cloud Diagnosis
di: Yang, Pei, et al.
Pubblicazione: (2026)

Learning without Forgetting for Vision-Language Models
di: Zhou, Da-Wei, et al.
Pubblicazione: (2023)

StutterFuse: Mitigating Modality Collapse in Stuttering Detection with Jaccard-Weighted Metric Learning and Gated Fusion
di: Singh, Guransh, et al.
Pubblicazione: (2025)

Understanding and Mitigating Miscalibration in Prompt Tuning for Vision-Language Models
di: Wang, Shuoyuan, et al.
Pubblicazione: (2024)

Gated Adaptation for Continual Learning in Human Activity Recognition
di: Azghan, Reza Rahimi, et al.
Pubblicazione: (2026)

When Counterfactual Reasoning Fails: Chaos and Real-World Complexity
di: Aalaila, Yahya, et al.
Pubblicazione: (2025)

Reliability-Gated Source Anchoring for Continual Test-Time Adaptation
di: Singh, Vikash, et al.
Pubblicazione: (2026)

MMLoP: Multi-Modal Low-Rank Prompting for Efficient Vision-Language Adaptation
di: Ghiasvand, Sajjad, et al.
Pubblicazione: (2026)