Concrete Jungle: Towards Concreteness Paved Contrastive Negative Mining for Compositional Understanding

• URL: http://arxiv.org/abs/2604.13313v1
• Date: Tue, 14 Apr 2026 21:28:46 GMT
• Title: Concrete Jungle: Towards Concreteness Paved Contrastive Negative Mining for Compositional Understanding
• Abstract summary: Vision-Language Models demonstrate remarkable capabilities but often struggle with compositional reasoning.<n>This study establishes lexical concreteness as a fundamental determinant of negative sample efficacy.<n>The integrated framework, designated as Slipform, achieves state-of-the-art accuracy across diverse compositional evaluation benchmarks.
• Score: 10.95036747579563
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Vision-Language Models demonstrate remarkable capabilities but often struggle with compositional reasoning, exhibiting vulnerabilities regarding word order and attribute binding. This limitation arises from a scarcity of informative samples needed to differentiate subtle semantic variations during contrastive pretraining. Although hard negative mining offers a promising remedy, existing methods lack explicit mechanisms to dictate which linguistic elements undergo modification. Instead of engineering generative architectures, this study establishes lexical concreteness as a fundamental determinant of negative sample efficacy. Modifying highly concrete terms generates more pronounced structural and visual discrepancies, providing a substantially stronger learning signal. Leveraging this principle, ConcretePlant is proposed to systematically isolate and manipulate perceptually grounded concepts. Analyses of the InfoNCE further reveals a severe gradient imbalance, where easily distinguishable pairs disproportionately overwhelm the optimization process and restrict the bandwidth available for nuanced learning. To resolve this degradation, the Cement loss is formulated utilizing a margin-based approach. By correlating psycholinguistic scores with sample difficulty, this objective dynamically calibrates the penalization applied to individual training pairs. Comprehensive evaluations substantiate these theoretical claims. The integrated framework, designated as Slipform, achieves state-of-the-art accuracy across diverse compositional evaluation benchmarks, general cross-modal retrieval, single and multi label linear probing.

Related papers

• Interpreting Object-Dependent Concept Brittleness in Text-to-Image Diffusion Models [105.53190946111586]
  We term this phenomenon object-dependent concept brittleness.<n>We present an interpretability-oriented framework to audit and minimally correct these failures.<n>We evaluate the proposed framework on style and attribute failure cases across multiple diffusion backbones.
  arXiv  Detail & Related papers  (2026-09-09T09:07:40Z)
• Rethinking Entropy Minimization in Test-Time Adaptation for Autoregressive Models [52.18157855348584]
  Test-Time Adaptation (TTA) via entropy minimization (EM) has proven effective for classification tasks, yet its application remains theoretically fragmented.<n>In this work, we deriving a rigorous formulation of EM tailored to autoregressive models.<n>We show that the exact objective naturally decomposes into a token-level policy gradient loss and a token-level entropy loss, and we reinterpret prior methods as partial realizations of this unified formulation.<n>Using Whisper ASR as a testbed, we demonstrate that our approach consistently improves performance across more than 20 diverse domains, including acoustic noise, accents, and multilingual settings.
  arXiv  Detail & Related papers  (2026-05-05T12:00:06Z)
• Delayed Homomorphic Reinforcement Learning for Environments with Delayed Feedback [11.866061471514582]
  Delayed feedback breaks the Markov assumption and impedes learning and control.<n>We propose a framework grounded in MDP homomorphisms that collapses belief-equivalent augmented states.<n> Experiments on continuous control tasks in MuJoCo benchmark confirm that our algorithm outperforms strong augmentation-based baselines.
  arXiv  Detail & Related papers  (2026-04-04T08:38:52Z)
• MSA-Thinker: Discrimination-Calibration Reasoning with Hint-Guided Reinforcement Learning for Multimodal Sentiment Analysis [5.1150258716324055]
  Multimodal sentiment analysis aims to understand human emotions by integrating textual, auditory, and visual modalities.<n>Existing methods incorporating Chain-of-Thought (CoT) reasoning are hindered by high annotation costs.<n>We propose a novel training framework that integrates structured Discrimination-Calibration (DC) reasoning with Hint-based Reinforcement Learning.
  arXiv  Detail & Related papers  (2026-03-10T12:48:41Z)
• Provable Benefit of Curriculum in Transformer Tree-Reasoning Post-Training [76.12556589212666]
  We show that curriculum post-training avoids the exponential complexity bottleneck.<n>Under outcome-only reward signals, reinforcement learning finetuning achieves high accuracy with sample complexity.<n>We establish guarantees for test-time scaling, where curriculum-aware querying reduces both reward oracle calls and sampling cost from exponential to order.
  arXiv  Detail & Related papers  (2025-11-10T18:29:54Z)
• Causality-aligned Prompt Learning via Diffusion-based Counterfactual Generation [45.395353088233556]
  We introduce a theoretically grounded $textbfDi$ffusion-based $textbfC$ounterf$textbfa$ctual $textbfp$rompt learning framework.<n>Our method performs excellently across tasks such as image classification, image-text retrieval, and visual question answering, with particularly strong advantages in unseen categories.
  arXiv  Detail & Related papers  (2025-07-26T09:27:52Z)
• Paying Alignment Tax with Contrastive Learning [6.232983467016873]
  Current debiasing approaches often result in a degradation in model capabilities such as factual accuracy and knowledge retention.<n>We propose a contrastive learning framework that learns through carefully constructed positive and negative examples.
  arXiv  Detail & Related papers  (2025-05-25T21:26:18Z)
• Regularized Neural Ensemblers [55.15643209328513]
  In this study, we explore employing regularized neural networks as ensemble methods.<n>Motivated by the risk of learning low-diversity ensembles, we propose regularizing the ensembling model by randomly dropping base model predictions.<n>We demonstrate this approach provides lower bounds for the diversity within the ensemble, reducing overfitting and improving generalization capabilities.
  arXiv  Detail & Related papers  (2024-10-06T15:25:39Z)
• A Unified Generalization Analysis of Re-Weighting and Logit-Adjustment for Imbalanced Learning [129.63326990812234]
  We propose a technique named data-dependent contraction to capture how modified losses handle different classes. On top of this technique, a fine-grained generalization bound is established for imbalanced learning, which helps reveal the mystery of re-weighting and logit-adjustment.
  arXiv  Detail & Related papers  (2023-10-07T09:15:08Z)
• Consistency Training with Virtual Adversarial Discrete Perturbation [17.311821099484987]
  We propose an effective consistency training framework that enforces a training model's predictions given original and perturbed inputs to be similar. This virtual adversarial discrete noise obtained by replacing a small portion of tokens efficiently pushes a training model's decision boundary.
  arXiv  Detail & Related papers  (2021-04-15T07:49:43Z)
• Deep Clustering by Semantic Contrastive Learning [67.28140787010447]
  We introduce a novel variant called Semantic Contrastive Learning (SCL) It explores the characteristics of both conventional contrastive learning and deep clustering. It can amplify the strengths of contrastive learning and deep clustering in a unified approach.
  arXiv  Detail & Related papers  (2021-03-03T20:20:48Z)
• Learning Causal Semantic Representation for Out-of-Distribution Prediction [125.38836464226092]
  We propose a Causal Semantic Generative model (CSG) based on a causal reasoning so that the two factors are modeled separately. We show that CSG can identify the semantic factor by fitting training data, and this semantic-identification guarantees the boundedness of OOD generalization error.
  arXiv  Detail & Related papers  (2020-11-03T13:16:05Z)
• Learning explanations that are hard to vary [75.30552491694066]
  We show that averaging across examples can favor memorization and patchwork' solutions that sew together different strategies. We then propose and experimentally validate a simple alternative algorithm based on a logical AND.
  arXiv  Detail & Related papers  (2020-09-01T10:17:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.