We introduce OmniRL, a highly generalizable in-context reinforcement learning (ICRL) model that is meta-trained on hundreds of thousands of diverse tasks. These tasks are procedurally generated by randomizing state transitions and rewards within Markov Decision Processes. To facilitate this extensive meta-training, we propose two key innovations: 1. An efficient data synthesis pipeline for ICRL, which leverages the interaction histories of diverse behavior policies; and 2. A novel modeling framework that integrates both imitation learning and reinforcement learning (RL) within the context, by incorporating prior knowledge. For the first time, we demonstrate that in-context learning (ICL) alone, without any gradient-based fine-tuning, can successfully tackle unseen Gymnasium tasks through imitation learning, online RL, or offline RL. Additionally, we show that achieving generalized ICRL capabilities-unlike task identification-oriented few-shot learning-critically depends on long trajectories generated by variant tasks and diverse behavior policies. By emphasizing the potential of ICL and departing from pre-training focused on acquiring specific skills, we further underscore the significance of meta-training aimed at cultivating the ability of ICL itself.

Robot picking and packing tasks require dexterous manipulation skills, such as rearranging objects to establish a good grasping pose, or placing and pushing items to achieve tight packing. These tasks are challenging for robots due to the complexity and variability of the required actions. To tackle the difficulty of learning and executing long-horizon tasks, we propose a novel framework called the Multi-Head Skill Transformer (MuST). This model is designed to learn and sequentially chain together multiple motion primitives (skills), enabling robots to perform complex sequences of actions effectively. MuST introduces a "progress value" for each skill, guiding the robot on which skill to execute next and ensuring smooth transitions between skills. Additionally, our model is capable of expanding its skill set and managing various sequences of sub-tasks efficiently. Extensive experiments in both simulated and real-world environments demonstrate that MuST significantly enhances the robot's ability to perform long-horizon dexterous manipulation tasks.

Dynamic pricing strategies are crucial for firms to maximize revenue by adjusting prices based on market conditions and customer characteristics. However, designing optimal pricing strategies becomes challenging when historical data are limited, as is often the case when launching new products or entering new markets. One promising approach to overcome this limitation is to leverage information from related products or markets to inform the focal pricing decisions. In this paper, we explore transfer learning for nonparametric contextual dynamic pricing under a covariate shift model, where the marginal distributions of covariates differ between source and target domains while the reward functions remain the same. We propose a novel Transfer Learning for Dynamic Pricing (TLDP) algorithm that can effectively leverage pre-collected data from a source domain to enhance pricing decisions in the target domain. The regret upper bound of TLDP is established under a simple Lipschitz condition on the reward function. To establish the optimality of TLDP, we further derive a matching minimax lower bound, which includes the target-only scenario as a special case and is presented for the first time in the literature. Extensive numerical experiments validate our approach, demonstrating its superiority over existing methods and highlighting its practical utility in real-world applications.

The o1 model series is trained with large-scale reinforcement learning to reason using chain of thought. These advanced reasoning capabilities provide new avenues for improving the safety and robustness of our models. In particular, our models can reason about our safety policies in context when responding to potentially unsafe prompts, through deliberative alignment. This leads to state-of-the-art performance on certain benchmarks for risks such as generating illicit advice, choosing stereotyped responses, and succumbing to known jailbreaks. Training models to incorporate a chain of thought before answering has the potential to unlock substantial benefits, while also increasing potential risks that stem from heightened intelligence. Our results underscore the need for building robust alignment methods, extensively stress-testing their efficacy, and maintaining meticulous risk management protocols. This report outlines the safety work carried out for the OpenAI o1 and OpenAI o1-mini models, including safety evaluations, external red teaming, and Preparedness Framework evaluations.

The increasing sizes of large language models (LLMs) result in significant computational overhead and memory usage when adapting these models to specific tasks or domains. Various parameter-efficient fine-tuning (PEFT) methods have been devised to mitigate these challenges by training a small set of parameters for the task-specific updates of the model weights. Among PEFT methods, LoRA stands out for its simplicity and efficiency, inspiring the development of a series of variants. However, LoRA and its successors disregard the knowledge that is noisy or irrelevant to the targeted task, detrimentally impacting model performance and leading to suboptimality. To address this limitation, we introduce Knowledge-aware Singular-value Adaptation (KaSA), a PEFT method that leverages singular value decomposition (SVD) with knowledge-aware singular values to dynamically activate knowledge based on its relevance to the task at hand. We conduct extensive experiments across a range of LLMs on tasks spanning natural language understanding (NLU), generation (NLG), instruction following, and commonsense reasoning. The experimental results demonstrate that KaSA consistently outperforms FFT and 14 popular PEFT baselines across 16 benchmarks and 4 synthetic datasets, underscoring our method's efficacy and adaptability. The source code of our method is available at https://github.com/juyongjiang/KaSA.

In domain adaptation, there are two popular paradigms: Unsupervised Domain Adaptation (UDA), which aligns distributions using source data, and Source-Free Domain Adaptation (SFDA), which leverages pre-trained source models without accessing source data. Evaluating the superiority of UDA versus SFDA is an open and timely question with significant implications for deploying adaptive algorithms in practical applications. In this study, we demonstrate through predictive coding theory and extensive experiments on multiple benchmark datasets that SFDA generally outperforms UDA in real-world scenarios. Specifically, SFDA offers advantages in time efficiency, storage requirements, targeted learning objectives, reduced risk of negative transfer, and increased robustness against overfitting. Notably, SFDA is particularly effective in mitigating negative transfer when there are substantial distribution discrepancies between source and target domains. Additionally, we introduce a novel data-model fusion scenario, where data sharing among stakeholders varies (e.g., some provide raw data while others provide only models), and reveal that traditional UDA and SFDA methods do not fully exploit their potential in this context. To address this limitation and capitalize on the strengths of SFDA, we propose a novel weight estimation method that effectively integrates available source data into multi-SFDA (MSFDA) approaches, thereby enhancing model performance within this scenario. This work provides a thorough analysis of UDA versus SFDA and advances a practical approach to model adaptation across diverse real-world environments.

As the prevalence of mental health challenges, social media has emerged as a key platform for individuals to express their emotions.Deep learning tends to be a promising solution for analyzing mental health on social media. However, black box models are often inflexible when switching between tasks, and their results typically lack explanations. With the rise of large language models (LLMs), their flexibility has introduced new approaches to the field. Also due to the generative nature, they can be prompted to explain decision-making processes. However, their performance on complex psychological analysis still lags behind deep learning. In this paper, we introduce the first multi-task Chinese Social Media Interpretable Mental Health Instructions (C-IMHI) dataset, consisting of 9K samples, which has been quality-controlled and manually validated. We also propose MentalGLM series models, the first open-source LLMs designed for explainable mental health analysis targeting Chinese social media, trained on a corpus of 50K instructions. The proposed models were evaluated on three downstream tasks and achieved better or comparable performance compared to deep learning models, generalized LLMs, and task fine-tuned LLMs. We validated a portion of the generated decision explanations with experts, showing promising results. We also evaluated the proposed models on a clinical dataset, where they outperformed other LLMs, indicating their potential applicability in the clinical field. Our models show strong performance, validated across tasks and perspectives. The decision explanations enhance usability and facilitate better understanding and practical application of the models. Both the constructed dataset and the models are publicly available via: https://github.com/zwzzzQAQ/MentalGLM.

Fine-tuning and in-context learning (ICL) are two prevalent methods in imbuing large language models with task-specific knowledge. It is commonly believed that fine-tuning can surpass ICL given sufficient training samples as it allows the model to adjust its internal parameters based on the data. However, this paper presents a counterintuitive finding: For tasks with implicit patterns, ICL captures these patterns significantly better than fine-tuning. We developed several datasets featuring implicit patterns, such as sequences determining answers through parity or identifying reducible terms in calculations. We then evaluated the models' understanding of these patterns under both fine-tuning and ICL across models ranging from 0.5B to 7B parameters. The results indicate that models employing ICL can quickly grasp deep patterns and significantly improve accuracy. In contrast, fine-tuning, despite utilizing thousands of times more training samples than ICL, achieved only limited improvements. We also proposed circuit shift theory from a mechanistic interpretability's view to explain why ICL wins.

Deep learning-based models are widely deployed in autonomous driving areas, especially the increasingly noticed end-to-end solutions. However, the black-box property of these models raises concerns about their trustworthiness and safety for autonomous driving, and how to debug the causality has become a pressing concern. Despite some existing research on the explainability of autonomous driving, there is currently no systematic solution to help researchers debug and identify the key factors that lead to the final predicted action of end-to-end autonomous driving. In this work, we propose a comprehensive approach to explore and analyze the causality of end-to-end autonomous driving. First, we validate the essential information that the final planning depends on by using controlled variables and counterfactual interventions for qualitative analysis. Then, we quantitatively assess the factors influencing model decisions by visualizing and statistically analyzing the response of key model inputs. Finally, based on the comprehensive study of the multi-factorial end-to-end autonomous driving system, we have developed a strong baseline and a tool for exploring causality in the close-loop simulator CARLA. It leverages the essential input sources to obtain a well-designed model, resulting in highly competitive capabilities. As far as we know, our work is the first to unveil the mystery of end-to-end autonomous driving and turn the black box into a white one. Thorough close-loop experiments demonstrate that our method can be applied to end-to-end autonomous driving solutions for causality debugging. Code will be available at https://github.com/bdvisl/DriveInsight.

World models are receiving increasing attention in autonomous driving for their ability to predict potential future scenarios. In this paper, we present BEVWorld, a novel approach that tokenizes multimodal sensor inputs into a unified and compact Bird's Eye View (BEV) latent space for environment modeling. The world model consists of two parts: the multi-modal tokenizer and the latent BEV sequence diffusion model. The multi-modal tokenizer first encodes multi-modality information and the decoder is able to reconstruct the latent BEV tokens into LiDAR and image observations by ray-casting rendering in a self-supervised manner. Then the latent BEV sequence diffusion model predicts future scenarios given action tokens as conditions. Experiments demonstrate the effectiveness of BEVWorld in autonomous driving tasks, showcasing its capability in generating future scenes and benefiting downstream tasks such as perception and motion prediction. Code will be available at https://github.com/zympsyche/BevWorld.