Can GPT tell us why these images are synthesized? Empowering Multimodal Large Language Models for Forensics

The rapid development of generative AI facilitates content creation and makes image manipulation easier and more difficult to detect. While multimodal Large Language Models (LLMs) have encoded rich world knowledge, they are not inherently tailored for combating AI-generated Content (AIGC) and struggle to comprehend local forgery details. In this work, we investigate the application of multimodal LLMs in forgery detection. We propose a framework capable of evaluating image authenticity, localizing tampered regions, providing evidence, and tracing generation methods based on semantic tampering clues. Our method demonstrates that the potential of LLMs in forgery analysis can be effectively unlocked through meticulous prompt engineering and the application of few-shot learning techniques. We conduct qualitative and quantitative experiments and show that GPT4V can achieve an accuracy of 92.1% in Autosplice and 86.3% in LaMa, which is competitive with state-of-the-art AIGC detection methods. We further discuss the limitations of multimodal LLMs in such tasks and propose potential improvements.

Command A: An Enterprise-Ready Large Language Model

In this report we describe the development of Command A, a powerful large language model purpose-built to excel at real-world enterprise use cases. Command A is an agent-optimised and multilingual-capable model, with support for 23 languages of global business, and a novel hybrid architecture balancing efficiency with top of the range performance. It offers best-in-class Retrieval Augmented Generation (RAG) capabilities with grounding and tool use to automate sophisticated business processes. These abilities are achieved through a decentralised training approach, including self-refinement algorithms and model merging techniques. We also include results for Command R7B which shares capability and architectural similarities to Command A. Weights for both models have been released for research purposes. This technical report details our original training pipeline and presents an extensive evaluation of our models across a suite of enterprise-relevant tasks and public benchmarks, demonstrating excellent performance and efficiency.

Language Models Can See Better: Visual Contrastive Decoding For LLM Multimodal Reasoning

Although Large Language Models (LLMs) excel in reasoning and generation for language tasks, they are not specifically designed for multimodal challenges. Training Multimodal Large Language Models (MLLMs), however, is resource-intensive and constrained by various training limitations. In this paper, we propose the Modular-based Visual Contrastive Decoding (MVCD) framework to move this obstacle. Our framework leverages LLMs' In-Context Learning (ICL) capability and the proposed visual contrastive-example decoding (CED), specifically tailored for this framework, without requiring any additional training. By converting visual signals into text and focusing on contrastive output distributions during decoding, we can highlight the new information introduced by contextual examples, explore their connections, and avoid over-reliance on prior encoded knowledge. MVCD enhances LLMs' visual perception to make it see and reason over the input visuals. To demonstrate MVCD's effectiveness, we conduct experiments with four LLMs across five question answering datasets. Our results not only show consistent improvement in model accuracy but well explain the effective components inside our decoding strategy. Our code will be available at https://github.com/Pbhgit/MVCD.

Rope to Nope and Back Again: A New Hybrid Attention Strategy

Long-context large language models (LLMs) have achieved remarkable advancements, driven by techniques like Rotary Position Embedding (RoPE) (Su et al., 2023) and its extensions (Chen et al., 2023; Liu et al., 2024c; Peng et al., 2023). By adjusting RoPE parameters and incorporating training data with extended contexts, we can train performant models with considerably longer input sequences. However, existing RoPE-based methods exhibit performance limitations when applied to extended context lengths. This paper presents a comprehensive analysis of various attention mechanisms, including RoPE, No Positional Embedding (NoPE), and Query-Key Normalization (QK-Norm), identifying their strengths and shortcomings in long-context modeling. Our investigation identifies distinctive attention patterns in these methods and highlights their impact on long-context performance, providing valuable insights for architectural design. Building on these findings, we propose a novel architectural based on a hybrid attention mechanism that not only surpasses conventional RoPE-based transformer models in long context tasks but also achieves competitive performance on benchmarks requiring shorter context lengths.

Mimicking-Bench: A Benchmark for Generalizable Humanoid-Scene Interaction Learning via Human Mimicking

Learning generic skills for humanoid robots interacting with 3D scenes by mimicking human data is a key research challenge with significant implications for robotics and real-world applications. However, existing methodologies and benchmarks are constrained by the use of small-scale, manually collected demonstrations, lacking the general dataset and benchmark support necessary to explore scene geometry generalization effectively. To address this gap, we introduce Mimicking-Bench, the first comprehensive benchmark designed for generalizable humanoid-scene interaction learning through mimicking large-scale human animation references. Mimicking-Bench includes six household full-body humanoid-scene interaction tasks, covering 11K diverse object shapes, along with 20K synthetic and 3K real-world human interaction skill references. We construct a complete humanoid skill learning pipeline and benchmark approaches for motion retargeting, motion tracking, imitation learning, and their various combinations. Extensive experiments highlight the value of human mimicking for skill learning, revealing key challenges and research directions.

Aya Expanse: Combining Research Breakthroughs for a New Multilingual Frontier

We introduce the Aya Expanse model family, a new generation of 8B and 32B parameter multilingual language models, aiming to address the critical challenge of developing highly performant multilingual models that match or surpass the capabilities of monolingual models. By leveraging several years of research at Cohere For AI and Cohere, including advancements in data arbitrage, multilingual preference training, and model merging, Aya Expanse sets a new state-of-the-art in multilingual performance. Our evaluations on the Arena-Hard-Auto dataset, translated into 23 languages, demonstrate that Aya Expanse 8B and 32B outperform leading open-weight models in their respective parameter classes, including Gemma 2, Qwen 2.5, and Llama 3.1, achieving up to a 76.6% win-rate. Notably, Aya Expanse 32B outperforms Llama 3.1 70B, a model with twice as many parameters, achieving a 54.0% win-rate. In this short technical report, we present extended evaluation results for the Aya Expanse model family and release their open-weights, together with a new multilingual evaluation dataset m-ArenaHard.

Real-Time Metric-Semantic Mapping for Autonomous Navigation in Outdoor Environments

The creation of a metric-semantic map, which encodes human-prior knowledge, represents a high-level abstraction of environments. However, constructing such a map poses challenges related to the fusion of multi-modal sensor data, the attainment of real-time mapping performance, and the preservation of structural and semantic information consistency. In this paper, we introduce an online metric-semantic mapping system that utilizes LiDAR-Visual-Inertial sensing to generate a global metric-semantic mesh map of large-scale outdoor environments. Leveraging GPU acceleration, our mapping process achieves exceptional speed, with frame processing taking less than 7ms, regardless of scenario scale. Furthermore, we seamlessly integrate the resultant map into a real-world navigation system, enabling metric-semantic-based terrain assessment and autonomous point-to-point navigation within a campus environment. Through extensive experiments conducted on both publicly available and self-collected datasets comprising 24 sequences, we demonstrate the effectiveness of our mapping and navigation methodologies. Code has been publicly released: https://github.com/gogojjh/cobra

ALR$^2$: A Retrieve-then-Reason Framework for Long-context Question Answering

The context window of large language models (LLMs) has been extended significantly in recent years. However, while the context length that the LLM can process has grown, the capability of the model to accurately reason over that context degrades noticeably. This occurs because modern LLMs often become overwhelmed by the vast amount of information in the context; when answering questions, the model must identify and reason over relevant evidence sparsely distributed throughout the text. To alleviate the challenge of long-context reasoning, we develop a retrieve-then-reason framework, enabling LLMs to reason over relevant evidence collected during an intermediate retrieval step. We find that modern LLMs struggle to accurately retrieve relevant facts and instead, often hallucinate "retrieved facts", resulting in flawed reasoning and the production of incorrect answers. To address these issues, we introduce ALR$^2$, a method that augments the long-context reasoning capability of LLMs via an explicit two-stage procedure, i.e., aligning LLMs with the objectives of both retrieval and reasoning. We demonstrate the efficacy of ALR$^2$ for mitigating performance degradation in long-context reasoning tasks. Through extensive experiments on long-context QA benchmarks, we find our method to outperform competitive baselines by large margins, achieving at least 8.4 and 7.9 EM gains on the long-context versions of HotpotQA and SQuAD datasets, respectively.

Uncertainty Quantification for Bird's Eye View Semantic Segmentation: Methods and Benchmarks

The fusion of raw features from multiple sensors on an autonomous vehicle to create a Bird's Eye View (BEV) representation is crucial for planning and control systems. There is growing interest in using deep learning models for BEV semantic segmentation. Anticipating segmentation errors and improving the explainability of DNNs is essential for autonomous driving, yet it is under-studied. This paper introduces a benchmark for predictive uncertainty quantification in BEV segmentation. The benchmark assesses various approaches across three popular datasets using two representative backbones and focuses on the effectiveness of predicted uncertainty in identifying misclassified and out-of-distribution (OOD) pixels, as well as calibration. Empirical findings highlight the challenges in uncertainty quantification. Our results find that evidential deep learning based approaches show the most promise by efficiently quantifying aleatoric and epistemic uncertainty. We propose the Uncertainty-Focal-Cross-Entropy (UFCE) loss, designed for highly imbalanced data, which consistently improves the segmentation quality and calibration. Additionally, we introduce a vacuity-scaled regularization term that enhances the model's focus on high uncertainty pixels, improving epistemic uncertainty quantification.

BeautyMap: Binary-Encoded Adaptable Ground Matrix for Dynamic Points Removal in Global Maps

Global point clouds that correctly represent the static environment features can facilitate accurate localization and robust path planning. However, dynamic objects introduce undesired ghost tracks that are mixed up with the static environment. Existing dynamic removal methods normally fail to balance the performance in computational efficiency and accuracy. In response, we present BeautyMap to efficiently remove the dynamic points while retaining static features for high-fidelity global maps. Our approach utilizes a binary-encoded matrix to efficiently extract the environment features. With a bit-wise comparison between matrices of each frame and the corresponding map region, we can extract potential dynamic regions. Then we use coarse to fine hierarchical segmentation of the $z$-axis to handle terrain variations. The final static restoration module accounts for the range-visibility of each single scan and protects static points out of sight. Comparative experiments underscore BeautyMap's superior performance in both accuracy and efficiency against other dynamic points removal methods. The code is open-sourced at https://github.com/MKJia/BeautyMap.