Sparse Attention Vectors: Generative Multimodal Model Features Are Discriminative Vision-Language Classifiers

Generative Large Multimodal Models (LMMs) like LLaVA and Qwen-VL excel at a wide variety of vision-language (VL) tasks such as image captioning or visual question answering. Despite strong performance, LMMs are not directly suited for foundational discriminative vision-language tasks (i.e., tasks requiring discrete label predictions) such as image classification and multiple-choice VQA. One key challenge in utilizing LMMs for discriminative tasks is the extraction of useful features from generative models. To overcome this issue, we propose an approach for finding features in the model's latent space to more effectively leverage LMMs for discriminative tasks. Toward this end, we present Sparse Attention Vectors (SAVs) -- a finetuning-free method that leverages sparse attention head activations (fewer than 1\% of the heads) in LMMs as strong features for VL tasks. With only few-shot examples, SAVs demonstrate state-of-the-art performance compared to a variety of few-shot and finetuned baselines on a collection of discriminative tasks. Our experiments also imply that SAVs can scale in performance with additional examples and generalize to similar tasks, establishing SAVs as both effective and robust multimodal feature representations.

Teaching VLMs to Localize Specific Objects from In-context Examples

Vision-Language Models (VLMs) have shown remarkable capabilities across diverse visual tasks, including image recognition, video understanding, and Visual Question Answering (VQA) when explicitly trained for these tasks. Despite these advances, we find that current VLMs lack a fundamental cognitive ability: learning to localize objects in a scene by taking into account the context. In this work, we focus on the task of few-shot personalized localization, where a model is given a small set of annotated images (in-context examples) -- each with a category label and bounding box -- and is tasked with localizing the same object type in a query image. To provoke personalized localization abilities in models, we present a data-centric solution that fine-tunes them using carefully curated data from video object tracking datasets. By leveraging sequences of frames tracking the same object across multiple shots, we simulate instruction-tuning dialogues that promote context awareness. To reinforce this, we introduce a novel regularization technique that replaces object labels with pseudo-names, ensuring the model relies on visual context rather than prior knowledge. Our method significantly enhances few-shot localization performance without sacrificing generalization, as demonstrated on several benchmarks tailored to personalized localization. This work is the first to explore and benchmark personalized few-shot localization for VLMs, laying a foundation for future research in context-driven vision-language applications. The code for our project is available at https://github.com/SivanDoveh/IPLoc

LiveXiv -- A Multi-Modal Live Benchmark Based on Arxiv Papers Content

The large-scale training of multi-modal models on data scraped from the web has shown outstanding utility in infusing these models with the required world knowledge to perform effectively on multiple downstream tasks. However, one downside of scraping data from the web can be the potential sacrifice of the benchmarks on which the abilities of these models are often evaluated. To safeguard against test data contamination and to truly test the abilities of these foundation models we propose LiveXiv: A scalable evolving live benchmark based on scientific ArXiv papers. LiveXiv accesses domain-specific manuscripts at any given timestamp and proposes to automatically generate visual question-answer pairs (VQA). This is done without any human-in-the-loop, using the multi-modal content in the manuscripts, like graphs, charts, and tables. Moreover, we introduce an efficient evaluation approach that estimates the performance of all models on the evolving benchmark using evaluations of only a subset of models. This significantly reduces the overall evaluation cost. We benchmark multiple open and proprietary Large Multi-modal Models (LMMs) on the first version of our benchmark, showing its challenging nature and exposing the models true abilities, avoiding contamination. Lastly, in our commitment to high quality, we have collected and evaluated a manually verified subset. By comparing its overall results to our automatic annotations, we have found that the performance variance is indeed minimal (<2.5%). Our dataset is available online on HuggingFace, and our code will be available here.

Augmenting In-Context-Learning in LLMs via Automatic Data Labeling and Refinement

It has been shown that Large Language Models' (LLMs) performance can be improved for many tasks using Chain of Thought (CoT) or In-Context Learning (ICL), which involve demonstrating the steps needed to solve a task using a few examples. However, while datasets with input-output pairs are relatively easy to produce, providing demonstrations which include intermediate steps requires cumbersome manual work. These steps may be executable programs, as in agentic flows, or step-by-step reasoning as in CoT. In this work, we propose Automatic Data Labeling and Refinement (ADLR), a method to automatically generate and filter demonstrations which include the above intermediate steps, starting from a small seed of manually crafted examples. We demonstrate the advantage of ADLR in code-based table QA and mathematical reasoning, achieving up to a 5.5% gain. The code implementing our method is provided in the Supplementary material and will be made available.

Multimodal Task Vectors Enable Many-Shot Multimodal In-Context Learning

The recent success of interleaved Large Multimodal Models (LMMs) in few-shot learning suggests that in-context learning (ICL) with many examples can be promising for learning new tasks. However, this many-shot multimodal ICL setting has one crucial problem: it is fundamentally limited by the model's context length set at pretraining. The problem is especially prominent in the multimodal domain, which processes both text and images, requiring additional tokens. This motivates the need for a multimodal method to compress many shots into fewer tokens without finetuning. In this work, we enable LMMs to perform multimodal, many-shot in-context learning by leveraging Multimodal Task Vectors (MTV)--compact implicit representations of in-context examples compressed in the model's attention heads. Specifically, we first demonstrate the existence of such MTV in LMMs and then leverage these extracted MTV to enable many-shot in-context learning for various vision-and-language tasks. Our experiments suggest that MTV can scale in performance with the number of compressed shots and generalize to similar out-of-domain tasks without additional context length for inference.

ConMe: Rethinking Evaluation of Compositional Reasoning for Modern VLMs

Compositional Reasoning (CR) entails grasping the significance of attributes, relations, and word order. Recent Vision-Language Models (VLMs), comprising a visual encoder and a Large Language Model (LLM) decoder, have demonstrated remarkable proficiency in such reasoning tasks. This prompts a crucial question: have VLMs effectively tackled the CR challenge? We conjecture that existing CR benchmarks may not adequately push the boundaries of modern VLMs due to the reliance on an LLM-only negative text generation pipeline. Consequently, the negatives produced either appear as outliers from the natural language distribution learned by VLMs' LLM decoders or as improbable within the corresponding image context. To address these limitations, we introduce ConMe -- a compositional reasoning benchmark and a novel data generation pipeline leveraging VLMs to produce `hard CR Q&A'. Through a new concept of VLMs conversing with each other to collaboratively expose their weaknesses, our pipeline autonomously generates, evaluates, and selects challenging compositional reasoning questions, establishing a robust CR benchmark, also subsequently validated manually. Our benchmark provokes a noteworthy, up to 33%, decrease in CR performance compared to preceding benchmarks, reinstating the CR challenge even for state-of-the-art VLMs.

NumeroLogic: Number Encoding for Enhanced LLMs' Numerical Reasoning

Language models struggle with handling numerical data and performing arithmetic operations. We hypothesize that this limitation can be partially attributed to non-intuitive textual numbers representation. When a digit is read or generated by a causal language model it does not know its place value (e.g. thousands vs. hundreds) until the entire number is processed. To address this issue, we propose a simple adjustment to how numbers are represented by including the count of digits before each number. For instance, instead of "42", we suggest using "{2:42}" as the new format. This approach, which we term NumeroLogic, offers an added advantage in number generation by serving as a Chain of Thought (CoT). By requiring the model to consider the number of digits first, it enhances the reasoning process before generating the actual number. We use arithmetic tasks to demonstrate the effectiveness of the NumeroLogic formatting. We further demonstrate NumeroLogic applicability to general natural language modeling, improving language understanding performance in the MMLU benchmark.

Towards Multimodal In-Context Learning for Vision & Language Models

Inspired by the emergence of Large Language Models (LLMs) that can truly understand human language, significant progress has been made in aligning other, non-language, modalities to be `understandable' by an LLM, primarily via converting their samples into a sequence of embedded language-like tokens directly fed into the LLM (decoder) input stream. However, so far limited attention has been given to transferring (and evaluating) one of the core LLM capabilities to the emerging VLMs, namely the In-Context Learning (ICL) ability, or in other words to guide VLMs to desired target downstream tasks or output structure using in-context image+text demonstrations. In this work, we dive deeper into analyzing the capabilities of some of the state-of-the-art VLMs to follow ICL instructions, discovering them to be somewhat lacking. We discover that even models that underwent large-scale mixed modality pre-training and were implicitly guided to make use of interleaved image and text information (intended to consume helpful context from multiple images) under-perform when prompted with few-shot (ICL) demonstrations, likely due to their lack of `direct' ICL instruction tuning. To test this conjecture, we propose a simple, yet surprisingly effective, strategy of extending a common VLM alignment framework with ICL support, methodology, and curriculum. We explore, analyze, and provide insights into effective data mixes, leading up to a significant 21.03% (and 11.3% on average) ICL performance boost over the strongest VLM baselines and a variety of ICL benchmarks. We also contribute new benchmarks for ICL evaluation in VLMs and discuss their advantages over the prior art.

Dense and Aligned Captions (DAC) Promote Compositional Reasoning in VL Models

Vision and Language (VL) models offer an effective method for aligning representation spaces of images and text, leading to numerous applications such as cross-modal retrieval, visual question answering, captioning, and more. However, the aligned image-text spaces learned by all the popular VL models are still suffering from the so-called `object bias' - their representations behave as `bags of nouns', mostly ignoring or downsizing the attributes, relations, and states of objects described/appearing in texts/images. Although some great attempts at fixing these `compositional reasoning' issues were proposed in the recent literature, the problem is still far from being solved. In this paper, we uncover two factors limiting the VL models' compositional reasoning performance. These two factors are properties of the paired VL dataset used for finetuning and pre-training the VL model: (i) the caption quality, or in other words `image-alignment', of the texts; and (ii) the `density' of the captions in the sense of mentioning all the details appearing on the image. We propose a fine-tuning approach for automatically treating these factors leveraging a standard VL dataset (CC3M). Applied to CLIP, we demonstrate its significant compositional reasoning performance increase of up to $\sim27\%$ over the base model, up to $\sim20\%$ over the strongest baseline, and by $6.7\%$ on average.

Incorporating Structured Representations into Pretrained Vision & Language Models Using Scene Graphs

Vision and Language (VL) models have demonstrated remarkable zero-shot performance in a variety of tasks. However, recent studies have shown that even the best VL models struggle to capture aspects of scene understanding, such as object attributes, relationships, and action states. In contrast, obtaining structured annotations, e.g., scene graphs (SGs) that could improve these models is time-consuming, costly, and tedious, and thus cannot be used on a large scale. Here we ask, can small datasets containing SG annotations provide sufficient information for enhancing structured understanding of VL models? We show that it is indeed possible to improve VL models using such data by utilizing a specialized model architecture and a new training paradigm. Our approach captures structure-related information for both the visual and textual encoders by directly supervising both components when learning from SG labels. We use scene graph supervision to generate fine-grained captions based on various graph augmentations highlighting different compositional aspects of the scene, and to predict SG information using an open vocabulary approach by adding special ``Adaptive SG tokens'' to the visual encoder. Moreover, we design a new adaptation technique tailored specifically to the SG tokens that allows better learning of the graph prediction task while still maintaining zero-shot capabilities. Our model shows strong performance improvements on the Winoground and VL-checklist datasets with only a mild degradation in zero-shot performance.