Humans have the ability to reason about geometric patterns in images and scenes from a young age. However, developing large multimodal models (LMMs) capable of similar reasoning remains a challenge, highlighting the need for robust evaluation methods to assess these capabilities. We introduce TurtleBench, a benchmark designed to evaluate LMMs' capacity to interpret geometric patterns -- given visual examples, textual instructions, or both -- and generate precise code outputs. Inspired by turtle geometry, a notion used to teach children foundational coding and geometric concepts, TurtleBench features tasks with patterned shapes that have underlying algorithmic logic. Our evaluation reveals that leading LMMs struggle significantly with these tasks, with GPT-4o achieving only 19\% accuracy on the simplest tasks and few-shot prompting only marginally improves their performance ($<2\%$). TurtleBench highlights the gap between human and AI performance in intuitive and visual geometrical understanding, setting the stage for future research in this area. TurtleBench stands as one of the few benchmarks to evaluate the integration of visual understanding and code generation capabilities in LMMs, setting the stage for future research. Code and Dataset for this paper is provided here: https://github.com/sinaris76/TurtleBench

Large Language Models (LLMs) have propelled groundbreaking advancements across several domains and are commonly used for text generation applications. However, the computational demands of these complex models pose significant challenges, requiring efficient hardware acceleration. Benchmarking the performance of LLMs across diverse hardware platforms is crucial to understanding their scalability and throughput characteristics. We introduce LLM-Inference-Bench, a comprehensive benchmarking suite to evaluate the hardware inference performance of LLMs. We thoroughly analyze diverse hardware platforms, including GPUs from Nvidia and AMD and specialized AI accelerators, Intel Habana and SambaNova. Our evaluation includes several LLM inference frameworks and models from LLaMA, Mistral, and Qwen families with 7B and 70B parameters. Our benchmarking results reveal the strengths and limitations of various models, hardware platforms, and inference frameworks. We provide an interactive dashboard to help identify configurations for optimal performance for a given hardware platform.

We present SegLLM, a novel multi-round interactive reasoning segmentation model that enhances LLM-based segmentation by exploiting conversational memory of both visual and textual outputs. By leveraging a mask-aware multimodal LLM, SegLLM re-integrates previous segmentation results into its input stream, enabling it to reason about complex user intentions and segment objects in relation to previously identified entities, including positional, interactional, and hierarchical relationships, across multiple interactions. This capability allows SegLLM to respond to visual and text queries in a chat-like manner. Evaluated on the newly curated MRSeg benchmark, SegLLM outperforms existing methods in multi-round interactive reasoning segmentation by over 20%. Additionally, we observed that training on multi-round reasoning segmentation data enhances performance on standard single-round referring segmentation and localization tasks, resulting in a 5.5% increase in cIoU for referring expression segmentation and a 4.5% improvement in Acc@0.5 for referring expression localization.

As an important component of data exploration and integration, Column Type Annotation (CTA) aims to label columns of a table with one or more semantic types. With the recent development of Large Language Models (LLMs), researchers have started to explore the possibility of using LLMs for CTA, leveraging their strong zero-shot capabilities. In this paper, we build on this promising work and improve on LLM-based methods for CTA by showing how to use a Knowledge Graph (KG) to augment the context information provided to the LLM. Our approach, called RACOON, combines both pre-trained parametric and non-parametric knowledge during generation to improve LLMs' performance on CTA. Our experiments show that RACOON achieves up to a 0.21 micro F-1 improvement compared against vanilla LLM inference.

The safe and effective deployment of Large Language Models (LLMs) involves a critical step called alignment, which ensures that the model's responses are in accordance with human preferences. Prevalent alignment techniques, such as DPO, PPO and their variants, align LLMs by changing the pre-trained model weights during a phase called post-training. While predominant, these post-training methods add substantial complexity before LLMs can be deployed. Inference-time alignment methods avoid the complex post-training step and instead bias the generation towards responses that are aligned with human preferences. The best-known inference-time alignment method, called Best-of-N, is as effective as the state-of-the-art post-training procedures. Unfortunately, Best-of-N requires vastly more resources at inference time than standard decoding strategies, which makes it computationally not viable. In this work, we introduce Speculative Rejection, a computationally-viable inference-time alignment algorithm. It generates high-scoring responses according to a given reward model, like Best-of-N does, while being between 16 to 32 times more computationally efficient.

Code metamorphism refers to a computer programming exercise wherein the program modifies its own code (partial or entire) consistently and automatically while retaining its core functionality. This technique is often used for online performance optimization and automated crash recovery in certain mission-critical applications. However, the technique has been misappropriated by malware creators to bypass signature-based detection measures instituted by anti-malware engines. However, current code mutation engines used by threat actors offer only a limited degree of mutation, which is frequently detectable via static code analysis. The advent of large language models (LLMs), such as ChatGPT 4.0 and Google Bard may lead to a significant evolution in this landscape. These models have demonstrated a level of algorithm comprehension and code synthesis capability that closely resembles human abilities. This advancement has sparked concerns among experts that such models could be exploited by threat actors to generate sophisticated metamorphic malware. This paper explores the potential of several prominent LLMs for software code mutation that may be used to reconstruct (with mutation) existing malware code bases or create new forms of embedded mutation engines for next-gen metamorphic malwares. In this work, we introduce a framework for creating self-testing program mutation engines based on LLM/Transformer-based models. The proposed framework serves as an essential tool in testing next-gen metamorphic malware detection engines.

Membership inference attacks (MIA) attempt to verify the membership of a given data sample in the training set for a model. MIA has become relevant in recent years, following the rapid development of large language models (LLM). Many are concerned about the usage of copyrighted materials for training them and call for methods for detecting such usage. However, recent research has largely concluded that current MIA methods do not work on LLMs. Even when they seem to work, it is usually because of the ill-designed experimental setup where other shortcut features enable "cheating." In this work, we argue that MIA still works on LLMs, but only when multiple documents are presented for testing. We construct new benchmarks that measure the MIA performances at a continuous scale of data samples, from sentences (n-grams) to a collection of documents (multiple chunks of tokens). To validate the efficacy of current MIA approaches at greater scales, we adapt a recent work on Dataset Inference (DI) for the task of binary membership detection that aggregates paragraph-level MIA features to enable MIA at document and collection of documents level. This baseline achieves the first successful MIA on pre-trained and fine-tuned LLMs.

Real-world machine learning systems often encounter model performance degradation due to distributional shifts in the underlying data generating process (DGP). Existing approaches to addressing shifts, such as concept drift adaptation, are limited by their reason-agnostic nature. By choosing from a pre-defined set of actions, such methods implicitly assume that the causes of model degradation are irrelevant to what actions should be taken, limiting their ability to select appropriate adaptations. In this paper, we propose an alternative paradigm to overcome these limitations, called self-healing machine learning (SHML). Contrary to previous approaches, SHML autonomously diagnoses the reason for degradation and proposes diagnosis-based corrective actions. We formalize SHML as an optimization problem over a space of adaptation actions to minimize the expected risk under the shifted DGP. We introduce a theoretical framework for self-healing systems and build an agentic self-healing solution H-LLM which uses large language models to perform self-diagnosis by reasoning about the structure underlying the DGP, and self-adaptation by proposing and evaluating corrective actions. Empirically, we analyze different components of H-LLM to understand why and when it works, demonstrating the potential of self-healing ML.

Large Language Models (LLMs) have demonstrated remarkable success across a wide range of language tasks, but their deployment on edge devices remains challenging due to the substantial memory requirements imposed by their large parameter sizes. Weight-only quantization presents a promising solution to reduce the memory footprint of LLMs. However, existing approaches primarily focus on integer-bit quantization, limiting their adaptability to fractional-bit quantization tasks and preventing the full utilization of available storage space on devices. In this paper, we introduce Channel-Wise Mixed-Precision Quantization (CMPQ), a novel mixed-precision quantization method that allocates quantization precision in a channel-wise pattern based on activation distributions. By assigning different precision levels to different weight channels, CMPQ can adapt to any bit-width constraint. CMPQ employs a non-uniform quantization strategy and incorporates two outlier extraction techniques that collaboratively preserve the critical information, thereby minimizing the quantization loss. Experiments on different sizes of LLMs demonstrate that CMPQ not only enhances performance in integer-bit quantization tasks but also achieves significant performance gains with a modest increase in memory usage. CMPQ thus represents an adaptive and effective approach to LLM quantization, offering substantial benefits across diverse device capabilities. Large Language Models (LLMs), trained on massive text corpora and containing up to hundreds of billions of parameters, have demonstrated exceptional performance across a wide range of language tasks (Brown, 2020; Chowdhery et al., 2023; Du et al., 2022; Hoffmann et al., 2022; Thoppilan et al., 2022; Touvron et al., 2023a).

Data collaboration via Data Clean Room offers value but raises privacy concerns, which can be addressed through synthetic data and multi-table synthesizers. Common multi-table synthesizers fail to perform when subjects occur repeatedly in both tables. This is an urgent yet unresolved problem, since having both tables with repeating subjects is common. To improve performance in this scenario, we present the DEREC 3-step pre-processing pipeline to generalize adaptability of multi-table synthesizers. We also introduce the SIMPRO 3-aspect evaluation metrics, which leverage conditional distribution and large-scale simultaneous hypothesis testing to provide comprehensive feedback on synthetic data fidelity at both column and table levels. Results show that using DEREC improves fidelity, and multi-table synthesizers outperform single-table counterparts in collaboration settings. Together, the DEREC-SIMPRO pipeline offers a robust solution for generalizing data collaboration, promoting a more efficient, data-driven society.