CS-CLIP is a vision‑language model that improves compositional reasoning by using scene graphs to identify compositional elements and create structured negative examples through selective masking. The approach retains only the most contradictory negatives, encouraging the model to depend on compositional structure instead of surface cues. CS-CLIP achieves state‑of‑the‑art performance on compositional reasoning benchmarks while maintaining strong cross‑modal retrieval and downstream visual reasoning capabilities with fewer training samples.
arXiv:2606. 26794v1 Announce Type: cross Abstract: CLIP and its variants are widely adopted visual backbones in multimodal systems, but their pretraining remains dominated by descriptive image-text alignment.
arXiv:2609.31456v1 Announce Type: new Abstract: Vision-language models (VLMs) often struggle with compositional reasoning tasks, but the reasons for this underperformance remain unclear. A common hyp...
arXiv:2606. 13288v1 Announce Type: cross Abstract: Contrastively trained vision-language models like CLIP, have made remarkable progress in learning joint image-text representations, but still face challenges in compositional understanding.
The paper introduces Auto-Comp, a fully automated, concept-driven pipeline that generates photorealistic compositional benchmarks for vision‑language models. Auto‑Comp creates paired Minimal and Contextual samples for each concept, enabling isolation of core binding abilities from visio‑linguistic complexity. Evaluations across 25 models reveal consistent failures in attribute and relational binding, with context helping relational tasks but hindering attribute tasks due to visual clutter.
Pretrained vision-language models such as CLIP excel at zero-shot recognition but often fail at compositionality, particularly attribute-object and relational structures. Recent studies mitigate this...
arXiv:2609.00709v1 Announce Type: cross Abstract: Large Vision-Language Models produce fluent image descriptions but offer limited semantic control: users cannot reliably specify whether captions sho...
CinematicVQA is a new benchmark for evaluating large vision‑language models on film‑grammar reasoning. It introduces the Cinematic Scene Graph, a structured representation linking filming techniques to perceptual effects and narrative functions, and tests models on tasks beyond low‑level technique recognition. The study finds a semantic gap where models excel at describing visuals but struggle to identify underlying techniques, and shows that fine‑tuning improves performance on narrative function and multi‑hop reasoning.
arXiv:2608. 07570v1 Announce Type: cross Abstract: Explainable aesthetic image cropping requires not only localizing a visually pleasing crop but also explaining why it is preferred.
MLLMCLIP introduces a heterogeneous distillation framework that transfers multimodal knowledge from a generative Multimodal Large Language Model (MLLM) teacher directly into a discriminative CLIP student, eliminating the need for synthetic hard negatives. The method uses an attention-based per-layer token selection and a CKA-based distillation loss to bridge architectural differences between the two models. As a result, MLLMCLIP achieves state‑of‑the‑art compositional accuracy and improves zero‑shot classification and image‑text retrieval performance.
Intelligent systems that act in the world require image understanding that is both comprehensive and spatially grounded. Current vision-language models (VLMs) can generate fluent and detailed image ca...
PSGAgent is an agentic framework that improves Multimodal Large Language Models’ performance on Panoptic Scene Graph generation by enabling dynamic visual perception. It introduces a think‑act‑observe loop with a Region Zoom‑In Tool for detailed local exploration and a Visual Highlight Tool for focused segmentation, reducing noise from irrelevant context. A two‑stage training pipeline—first building a high‑quality, error‑aware Multimodal Chain‑of‑Thought dataset, then applying a hierarchical reward mechanism with Group Sequence Policy Optimization—optimizes tool usage and achieves state‑of‑the‑art results on the PSG benchmark.
Flow-CLIP introduces a method to improve compositional image-text matching by refining global image-text alignment with image-conditioned changes in token–patch evidence. It combines Global Compatibility Calibration (GCC) for sentence-level similarity, Image-Conditioned Evidence Refinement (ICER) to incorporate visual patch context, and Cross-Image Margin Ranking (CIMR) to enhance ranking margins when images support correct descriptions. Experiments demonstrate that Flow-CLIP outperforms pretrained CLIP and other global alignment approaches on binding accuracy and reduces reliance on language priors.