Vision-language models (VLMs) have made significant progress in recent visual-question-answering (VQA) benchmarks that evaluate complex visio-linguistic reasoning. However, are these models truly effective? In this work, we show that VLMs still struggle with natural images and questions that humans can easily answer, which we term $\textbf{natural adversarial samples}$. We also find it surprisingly easy to generate these VQA samples from natural image-text corpora using off-the-shelf models like CLIP and ChatGPT. We propose a semi-automated approach to collect a new benchmark, ${\bf NaturalBench}$, for reliably evaluating VLMs with 10,000 human-verified VQA samples.

Zero-shot reinforcement learning (RL) promises to provide agents that can perform task in an environment after an offline, reward-free pre-training phase. Methods leveraging successor measures and successor features have shown strong performance in this setting, but require access to large heterogenous datasets for pre-training which cannot be expected for most real problems. Here, we explore how the performance of zero-shot RL methods degrades when trained on small homogeneous datasets, and propose fixes inspired by, a well-established feature of performant single-task offline RL algorithms. We evaluate our proposals across various datasets, domains and tasks, and show that conservative zero-shot RL algorithms outperform their non-conservative counterparts on low quality datasets, and perform no worse on high quality datasets. Somewhat surprisingly, our proposals also outperform baselines that get to see the task during training.

As the adoption of large language models increases and the need for per-user or per-task model customization grows, the parameter-efficient fine-tuning (PEFT) methods, such as low-rank adaptation (LoRA) and its variants, incur substantial storage and transmission costs. To further reduce stored parameters, we introduce a divide-and-share paradigm that breaks the barriers of low-rank decomposition across matrix dimensions, modules, and layers by sharing parameters globally via a vector bank. As an instantiation of the paradigm to LoRA, our proposed VB-LoRA composites all the low-rank matrices of LoRA from a shared vector bank with a differentiable top-$k$ admixture module. VB-LoRA achieves extreme parameter efficiency while maintaining comparable or better performance compared to state-of-the-art PEFT methods. Extensive experiments demonstrate the effectiveness of VB-LoRA on natural language understanding, natural language generation, instruction tuning, and mathematical reasoning tasks. When fine-tuning the Llama2-13B model, VB-LoRA only uses 0.4% of LoRA's stored parameters, yet achieves superior results.

Inference on large language models (LLMs) can be expensive in terms of thecompute and memory costs involved, especially when long sequence lengths areused. In particular, the self-attention mechanism used in LLM inference contributessignificantly to these costs, which has sparked an interest in approximating the self-attention computation to reduce such costs. In this work, we propose to approximateself-attention by focusing on the dimensionality of key vectors computed in theattention block. Our analysis reveals that key vectors lie in a significantly lower-dimensional space, consistently across several datasets and models. Exploiting thisobservation, we propose Loki, a novel sparse attention method that ranks and selectstokens in the KV-cache based on attention scores computed in low-dimensionalspace. Our evaluations show that Loki is able to speed up the attention computationdue to reduced data movement (load/store) and compute costs while maintainingthe efficacy of the models better than other popular approximation methods.

Amidst the recent strides in evaluating Large Language Models for Code (Code LLMs), existing benchmarks have mainly focused on the functional correctness of generated code, neglecting the importance of their computational efficiency. To fill the gap, we present Mercury, the first code efficiency benchmark for Code LLMs. It comprises 1,889 Python tasks, each accompanied by adequate solutions that serve as real-world efficiency baselines, enabling a comprehensive analysis of the runtime distribution. Based on the distribution, we introduce a new metric Beyond, which computes a runtime-percentile-weighted Pass score to reflect functional correctness and code efficiency simultaneously. On Mercury, leading Code LLMs can achieve 65% on Pass, while less than 50% on Beyond. Given that an ideal Beyond score would be aligned with the Pass score, it indicates that while Code LLMs exhibit impressive capabilities in generating functionally correct code, there remains a notable gap in their efficiency. Finally, our empirical experiments reveal that Direct Preference Optimization (DPO) serves as a robust baseline for enhancing code efficiency compared with Supervised Fine Tuning (SFT), which paves a promising avenue for future exploration of efficient code generation. Our code and data are available on GitHub: https://github.com/Elfsong/Mercury.

Large language models (LLMs) have shown impressive performance on downstream tasks by in-context learning (ICL), which heavily relies on the quality of demonstrations selected from a large set of annotated examples. Recent works claim that in-context learning is robust to noisy demonstrations in text classification. In this work, we show that, on text generation tasks, noisy annotations significantly hurt the performance of in-context learning. To circumvent the issue, we propose a simple and effective approach called Local Perplexity Ranking (LPR), which replaces the noisy candidates with their nearest neighbors that are more likely to be clean. Our method is motivated by analyzing the perplexity deviation caused by noisy labels and decomposing perplexity into inherent perplexity and matching perplexity. Our key idea behind LPR is thus to decouple the matching perplexity by performing the ranking among the neighbors in semantic space. Our approach can prevent the selected demonstrations from including mismatched input-label pairs while preserving the effectiveness of the original selection methods. Extensive experiments demonstrate the effectiveness of LPR, improving the EM score by up to 18.75 on common benchmarks with noisy annotations.

Large language models (LLMs) like transformers demonstrate impressive in-context learning (ICL) capabilities, allowing them to makepredictions for new tasks based on prompt exemplars without parameter updates. While existing ICL theories often assume structured training data resembling ICL tasks (e.g., x-y pairs for linear regression), LLMs are typically trained unsupervised on unstructured text, such as web content, which lacks clear parallels to tasks like word analogy. To address this gap, we examine what enables ICL in models trained on unstructured data, focusing on critical sequence model requirements and training data structure. We find that many ICL capabilities canemerge simply from co-occurrence of semantically related word pairs in unstructured data; word analogy completion, for example, can provably arise purely through co-occurrence modeling, using classical language models like continuous bag of words (CBOW), without needing positional information or attention mechanisms. However, positional information becomes crucial for logic reasoning tasks requiring generalization to unseen tokens. Finally, we identify two cases where ICL fails: one in logic reasoning tasks that require generalizing to new, unseen patterns, and another in analogy completion where relevant word pairs appear only in fixed training positions. These findings suggest that LLMs' ICL abilities depend heavily on the structural elements within their training data.

As large language models gain widespread adoption, running them efficiently becomes a crucial task. Recent works on LLM inference use speculative decoding to achieve extreme speedups. However, most of these works implicitly design their algorithms for high-end datacenter hardware. In this work, we ask the opposite question: how fast can we run LLMs on consumer machines? Consumer GPUs can no longer fit the largest available models and must offload them to RAM or SSD. With parameter offloading, hundreds or thousands of tokens can be processed in batches within the same time as just one token, making it a natural fit for speculative decoding. We propose SpecExec (Speculative Execution), a simple parallel decoding method that can generate up to 20 tokens per target model iteration for popular LLM families. SpecExec takes the most probable continuations from the draft model to build a cache tree for the target model, which then gets validated in a single pass. Using SpecExec, we demonstrate inference of 50B+ parameter LLMs on consumer GPUs with RAM offloading at 4--6 tokens per second with 4-bit quantization or 2--3 tokens per second with 16-bit weights.

Large language models (LLMs) have the potential to generate texts that pose risks of misuse, such as plagiarism, planting fake reviews on e-commerce platforms, or creating inflammatory false tweets. Consequently, detecting whether a text is generated by LLMs has become increasingly important. Existing high-quality detection methods usually require access to the interior of the model to extract the intrinsic characteristics. However, since we do not have access to the interior of the black-box model, we must resort to surrogate models, which impacts detection quality. In order to achieve high-quality detection of black-box models, we would like to extract deep intrinsic characteristics of the black-box model generated texts.

PCWorld reports that Google's Gemini may soon receive conversation branching functionality, a feature currently unique to ChatGPT among major AI chatbots. This capability allows users to explore different conversational paths from any point without losing the original thread, enhancing experimentation and control. Android Authority discovered hints of this upcoming feature in Gemini's code, while competitors like Claude still lack branching functionality. Ever wish you could take an existing AI conversation in an entirely new directory while keeping the original chat thread intact? ChatGPT makes it easy with its "branching" feature, but Claude and Gemini don't offer any branching functionality-or at least, not yet.