Copertina

The Scaffold Effect: How Prompt Framing Drives Apparent Multimodal Gains in Clinical VLM Evaluation

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Vu, Doan Nam Long, Balloccu, Simone
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Artificial Intelligence Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Venn Diagram Prompting : Accelerating Comprehension with Scaffolding Effect
di: Mahendru, Sakshi, et al.
Pubblicazione: (2024)

How Prompts Move Language Model Behavior: Frames, Salience, and Construal as Semantic Control
di: Kim, Dongseok, et al.
Pubblicazione: (2025)

DatBench: Discriminative, Faithful, and Efficient VLM Evaluations
di: DatologyAI, et al.
Pubblicazione: (2026)

From Demonstrations to Rewards: Test-Time Prompt Optimization for VLM Reward Models
di: Gumbsch, Christian, et al.
Pubblicazione: (2026)

Faithful Path Language Modeling for Explainable Recommendation over Knowledge Graph
di: Balloccu, Giacomo, et al.
Pubblicazione: (2023)

No Prompt Left Behind: Exploiting Zero-Variance Prompts in LLM Reinforcement Learning via Entropy-Guided Advantage Shaping
di: Le, Thanh-Long V., et al.
Pubblicazione: (2025)

NDST: Neural Driving Style Transfer for Human-Like Vision-Based Autonomous Driving
di: Kim, Donghyun, et al.
Pubblicazione: (2024)

Multimodal Large Language Models for End-to-End Affective Computing: Benchmarking and Boosting with Generative Knowledge Prompting
di: Luo, Miaosen, et al.
Pubblicazione: (2025)

VLM-Guided Experience Replay
di: Sharony, Elad, et al.
Pubblicazione: (2026)

Learning Dynamics of VLM Finetuning
di: Zhang, Jusheng, et al.
Pubblicazione: (2025)

Safety Mirage: How Spurious Correlations Undermine VLM Safety Fine-Tuning and Can Be Mitigated by Machine Unlearning
di: Chen, Yiwei, et al.
Pubblicazione: (2025)

The Pragmatic Frames of Spurious Correlations in Machine Learning: Interpreting How and Why They Matter
di: Bell, Samuel J., et al.
Pubblicazione: (2024)

Separable Pathways for Causal Reasoning: How Architectural Scaffolding Enables Hypothesis-Space Restructuring in LLM Agents
di: Alderete, John, et al.
Pubblicazione: (2026)

BadPromptFL: A Novel Backdoor Threat to Prompt-based Federated Learning in Multimodal Models
di: Zhang, Maozhen, et al.
Pubblicazione: (2025)

LIBMoE: A Library for comprehensive benchmarking Mixture of Experts in Large Language Models
di: Nguyen, Nam V., et al.
Pubblicazione: (2024)

NeuroVLM-Bench: Evaluation of Vision-Enabled Large Language Models for Clinical Reasoning in Neurological Disorders
di: Dineva, Katarina Trojachanec, et al.
Pubblicazione: (2026)

MetaSynth: Meta-Prompting-Driven Agentic Scaffolds for Diverse Synthetic Data Generation
di: Riaz, Haris, et al.
Pubblicazione: (2025)

V2X-VLM: End-to-End V2X Cooperative Autonomous Driving Through Large Vision-Language Models
di: You, Junwei, et al.
Pubblicazione: (2024)

SafeCoT: Improving VLM Safety with Minimal Reasoning
di: Ma, Jiachen, et al.
Pubblicazione: (2025)

Quantifying the Gain in Weak-to-Strong Generalization
di: Charikar, Moses, et al.
Pubblicazione: (2024)

Measuring Cross-Modal Synergy: A Benchmark for VLM Explainability
di: Ky, Joël Roman, et al.
Pubblicazione: (2026)

FuzzingRL: Reinforcement Fuzz-Testing for Revealing VLM Failures
di: Xu, Jiajun, et al.
Pubblicazione: (2026)

Offline RLAIF: Piloting VLM Feedback for RL via SFO
di: Beck, Jacob
Pubblicazione: (2025)

CF-VLM:CounterFactual Vision-Language Fine-tuning
di: Zhang, Jusheng, et al.
Pubblicazione: (2025)

Soft Prompts for Evaluation: Measuring Conditional Distance of Capabilities
di: Nordby, Ross
Pubblicazione: (2025)

Evaluating Generalization and Representation Stability in Small LMs via Prompting, Fine-Tuning and Out-of-Distribution Prompts
di: Raja, Rahul, et al.
Pubblicazione: (2025)

MP-MoE: Matrix Profile-Guided Mixture of Experts for Precipitation Forecasting
di: Tran, Huyen Ngoc, et al.
Pubblicazione: (2026)

Maximizing Mutual Information Between Prompt and Response Improves LLM Performance With No Additional Data
di: Nam, Hyunji, et al.
Pubblicazione: (2026)

Promptception: How Sensitive Are Large Multimodal Models to Prompts?
di: Ismithdeen, Mohamed Insaf, et al.
Pubblicazione: (2025)

XGeM: A Multi-Prompt Foundation Model for Multimodal Medical Data Generation
di: Molino, Daniele, et al.
Pubblicazione: (2025)

How Susceptible are LLMs to Influence in Prompts?
di: Anagnostidis, Sotiris, et al.
Pubblicazione: (2024)

How Not to Detect Prompt Injections with an LLM
di: Choudhary, Sarthak, et al.
Pubblicazione: (2025)

MOSAIC: A Unified Platform for Cross-Paradigm Comparison and Evaluation of Homogeneous and Heterogeneous Multi-Agent RL, LLM, VLM, and Human Decision-Makers
di: Mousa, Abdulhamid M., et al.
Pubblicazione: (2026)

How Well Do Multimodal Models Reason on ECG Signals?
di: Xu, Maxwell A., et al.
Pubblicazione: (2026)

CLIP the Bias: How Useful is Balancing Data in Multimodal Learning?
di: Alabdulmohsin, Ibrahim, et al.
Pubblicazione: (2024)

Batch Normalization Amplifies Memorization and Privacy Risks
di: Doan, Ngoc Phu, et al.
Pubblicazione: (2026)

A Discordance-Aware Multimodal Framework with Multi-Agent Clinical Reasoning
di: Ahadian, Pegah, et al.
Pubblicazione: (2026)

Beyond Interpretability: The Gains of Feature Monosemanticity on Model Robustness
di: Zhang, Qi, et al.
Pubblicazione: (2024)

MTBBench: A Multimodal Sequential Clinical Decision-Making Benchmark in Oncology
di: Vasilev, Kiril, et al.
Pubblicazione: (2025)

GROW: Aligning GRPO with State-Action Modeling for Open-World VLM Agents
di: Wu, Xiongbin, et al.
Pubblicazione: (2026)