Redefining Superalignment: From Weak-to-Strong Alignment to Human-AI Co-Alignment to Sustainable Symbiotic Society

Artificial Intelligence (AI) systems are becoming increasingly powerful and autonomous, and may progress to surpass human intelligence levels, namely Artificial Superintelligence (ASI). During the progression from AI to ASI, it may exceed human control, violate human values, and even lead to irreversible catastrophic consequences in extreme cases. This gives rise to a pressing issue that needs to be addressed: superalignment, ensuring that AI systems much smarter than humans, remain aligned with human (compatible) intentions and values. Existing scalable oversight and weak-to-strong generalization methods may prove substantially infeasible and inadequate when facing ASI. We must explore safer and more pluralistic frameworks and approaches for superalignment. In this paper, we redefine superalignment as the human-AI co-alignment towards a sustainable symbiotic society, and highlight a framework that integrates external oversight and intrinsic proactive alignment. External oversight superalignment should be grounded in human-centered ultimate decision, supplemented by interpretable automated evaluation and correction, to achieve continuous alignment with humanity's evolving values. Intrinsic proactive superalignment is rooted in a profound understanding of the self, others, and society, integrating self-awareness, self-reflection, and empathy to spontaneously infer human intentions, distinguishing good from evil and proactively considering human well-being, ultimately attaining human-AI co-alignment through iterative interaction. The integration of externally-driven oversight with intrinsically-driven proactive alignment empowers sustainable symbiotic societies through human-AI co-alignment, paving the way for achieving safe and beneficial AGI and ASI for good, for human, and for a symbiotic ecology.

A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment

The remarkable success of Large Language Models (LLMs) has illuminated a promising pathway toward achieving Artificial General Intelligence for both academic and industrial communities, owing to their unprecedented performance across various applications. As LLMs continue to gain prominence in both research and commercial domains, their security and safety implications have become a growing concern, not only for researchers and corporations but also for every nation. Currently, existing surveys on LLM safety primarily focus on specific stages of the LLM lifecycle, e.g., deployment phase or fine-tuning phase, lacking a comprehensive understanding of the entire "lifechain" of LLMs. To address this gap, this paper introduces, for the first time, the concept of "full-stack" safety to systematically consider safety issues throughout the entire process of LLM training, deployment, and eventual commercialization. Compared to the off-the-shelf LLM safety surveys, our work demonstrates several distinctive advantages: (I) Comprehensive Perspective. We define the complete LLM lifecycle as encompassing data preparation, pre-training, post-training, deployment and final commercialization. To our knowledge, this represents the first safety survey to encompass the entire lifecycle of LLMs. (II) Extensive Literature Support. Our research is grounded in an exhaustive review of over 800+ papers, ensuring comprehensive coverage and systematic organization of security issues within a more holistic understanding. (III) Unique Insights. Through systematic literature analysis, we have developed reliable roadmaps and perspectives for each chapter. Our work identifies promising research directions, including safety in data generation, alignment techniques, model editing, and LLM-based agent systems. These insights provide valuable guidance for researchers pursuing future work in this field.

Benchmarking Multi-National Value Alignment for Large Language Models

Do Large Language Models (LLMs) hold positions that conflict with your country's values? Occasionally they do! However, existing works primarily focus on ethical reviews, failing to capture the diversity of national values, which encompass broader policy, legal, and moral considerations. Furthermore, current benchmarks that rely on spectrum tests using manually designed questionnaires are not easily scalable. To address these limitations, we introduce NaVAB, a comprehensive benchmark to evaluate the alignment of LLMs with the values of five major nations: China, the United States, the United Kingdom, France, and Germany. NaVAB implements a national value extraction pipeline to efficiently construct value assessment datasets. Specifically, we propose a modeling procedure with instruction tagging to process raw data sources, a screening process to filter value-related topics and a generation process with a Conflict Reduction mechanism to filter non-conflicting values.We conduct extensive experiments on various LLMs across countries, and the results provide insights into assisting in the identification of misaligned scenarios. Moreover, we demonstrate that NaVAB can be combined with alignment techniques to effectively reduce value concerns by aligning LLMs' values with the target country.

Dexterous Non-Prehensile Manipulation for Ungraspable Object via Extrinsic Dexterity

Objects with large base areas become ungraspable when they exceed the end-effector's maximum aperture. Existing approaches address this limitation through extrinsic dexterity, which exploits environmental features for non-prehensile manipulation. While grippers have shown some success in this domain, dexterous hands offer superior flexibility and manipulation capabilities that enable richer environmental interactions, though they present greater control challenges. Here we present ExDex, a dexterous arm-hand system that leverages reinforcement learning to enable non-prehensile manipulation for grasping ungraspable objects. Our system learns two strategic manipulation sequences: relocating objects from table centers to edges for direct grasping, or to walls where extrinsic dexterity enables grasping through environmental interaction. We validate our approach through extensive experiments with dozens of diverse household objects, demonstrating both superior performance and generalization capabilities with novel objects. Furthermore, we successfully transfer the learned policies from simulation to a real-world robot system without additional training, further demonstrating its applicability in real-world scenarios. Project website: https://tangty11.github.io/ExDex/.

ThinkPatterns-21k: A Systematic Study on the Impact of Thinking Patterns in LLMs

Large language models (LLMs) have demonstrated enhanced performance through the \textit{Thinking then Responding} paradigm, where models generate internal thoughts before final responses (aka, System 2 thinking). However, existing research lacks a systematic understanding of the mechanisms underlying how thinking patterns affect performance across model sizes. In this work, we conduct a comprehensive analysis of the impact of various thinking types on model performance and introduce ThinkPatterns-21k, a curated dataset comprising 21k instruction-response pairs (QA) collected from existing instruction-following datasets with five thinking types. For each pair, we augment it with five distinct internal thinking patterns: one unstructured thinking (monologue) and four structured variants (decomposition, self-ask, self-debate and self-critic), while maintaining the same instruction and response. Through extensive evaluation across different model sizes (3B-32B parameters), we have two key findings: (1) smaller models (<30B parameters) can benefit from most of structured thinking patterns, while larger models (32B) with structured thinking like decomposition would degrade performance and (2) unstructured monologue demonstrates broad effectiveness across different model sizes. Finally, we released all of our datasets, checkpoints, training logs of diverse thinking patterns to reproducibility, aiming to facilitate further research in this direction.

SafeVLA: Towards Safety Alignment of Vision-Language-Action Model via Safe Reinforcement Learning

Vision-language-action models (VLAs) have shown great potential as generalist robot policies. However, these models pose urgent safety challenges during deployment, including the risk of physical harm to the environment, the robot itself, and humans. How can safety be explicitly incorporated into VLAs? In this work, we propose SafeVLA, a novel algorithm designed to integrate safety into VLAs, ensuring the protection of the environment, robot hardware and humans in real-world settings. SafeVLA effectively balances safety and task performance by employing large-scale constrained learning within simulated environments. We demonstrate that SafeVLA outperforms the current state-of-the-art method in both safety and task performance, achieving average improvements of 83.58% and 3.85%, respectively, in simulation. By prioritizing safety, our approach eliminates high-risk behaviors and reduces the upper bound of unsafe behaviors to 1/35 of that in the current state-of-the-art, thereby significantly mitigating long-tail risks. Furthermore, the learned safety constraints generalize to diverse, unseen scenarios, including multiple out-of-distribution perturbations and tasks. Our data, models and newly proposed benchmark environment are available at https://sites.google.com/view/pku-safevla.

Differentiable Information Enhanced Model-Based Reinforcement Learning

Differentiable environments have heralded new possibilities for learning control policies by offering rich differentiable information that facilitates gradient-based methods. In comparison to prevailing model-free reinforcement learning approaches, model-based reinforcement learning (MBRL) methods exhibit the potential to effectively harness the power of differentiable information for recovering the underlying physical dynamics. However, this presents two primary challenges: effectively utilizing differentiable information to 1) construct models with more accurate dynamic prediction and 2) enhance the stability of policy training. In this paper, we propose a Differentiable Information Enhanced MBRL method, MB-MIX, to address both challenges. Firstly, we adopt a Sobolev model training approach that penalizes incorrect model gradient outputs, enhancing prediction accuracy and yielding more precise models that faithfully capture system dynamics. Secondly, we introduce mixing lengths of truncated learning windows to reduce the variance in policy gradient estimation, resulting in improved stability during policy learning. To validate the effectiveness of our approach in differentiable environments, we provide theoretical analysis and empirical results. Notably, our approach outperforms previous model-based and model-free methods, in multiple challenging tasks involving controllable rigid robots such as humanoid robots' motion control and deformable object manipulation.

DexGraspVLA: A Vision-Language-Action Framework Towards General Dexterous Grasping

Dexterous grasping remains a fundamental yet challenging problem in robotics. A general-purpose robot must be capable of grasping diverse objects in arbitrary scenarios. However, existing research typically relies on specific assumptions, such as single-object settings or limited environments, leading to constrained generalization. Our solution is DexGraspVLA, a hierarchical framework that utilizes a pre-trained Vision-Language model as the high-level task planner and learns a diffusion-based policy as the low-level Action controller. The key insight lies in iteratively transforming diverse language and visual inputs into domain-invariant representations, where imitation learning can be effectively applied due to the alleviation of domain shift. Thus, it enables robust generalization across a wide range of real-world scenarios. Notably, our method achieves a 90+% success rate under thousands of unseen object, lighting, and background combinations in a ``zero-shot'' environment. Empirical analysis further confirms the consistency of internal model behavior across environmental variations, thereby validating our design and explaining its generalization performance. We hope our work can be a step forward in achieving general dexterous grasping. Our demo and code can be found at https://dexgraspvla.github.io/.

Retrieval Dexterity: Efficient Object Retrieval in Clutters with Dexterous Hand

Retrieving objects buried beneath multiple objects is not only challenging but also time-consuming. Performing manipulation in such environments presents significant difficulty due to complex contact relationships. Existing methods typically address this task by sequentially grasping and removing each occluding object, resulting in lengthy execution times and requiring impractical grasping capabilities for every occluding object. In this paper, we present a dexterous arm-hand system for efficient object retrieval in multi-object stacked environments. Our approach leverages large-scale parallel reinforcement learning within diverse and carefully designed cluttered environments to train policies. These policies demonstrate emergent manipulation skills (e.g., pushing, stirring, and poking) that efficiently clear occluding objects to expose sufficient surface area of the target object. We conduct extensive evaluations across a set of over 10 household objects in diverse clutter configurations, demonstrating superior retrieval performance and efficiency for both trained and unseen objects. Furthermore, we successfully transfer the learned policies to a real-world dexterous multi-fingered robot system, validating their practical applicability in real-world scenarios. Videos can be found on our project website https://ChangWinde.github.io/RetrDex.

Amulet: ReAlignment During Test Time for Personalized Preference Adaptation of LLMs

How to align large language models (LLMs) with user preferences from a static general dataset has been frequently studied. However, user preferences are usually personalized, changing, and diverse regarding culture, values, or time. This leads to the problem that the actual user preferences often do not coincide with those trained by the model developers in the practical use of LLMs. Since we cannot collect enough data and retrain for every demand, researching efficient real-time preference adaptation methods based on the backbone LLMs during test time is important. To this end, we introduce Amulet, a novel, training-free framework that formulates the decoding process of every token as a separate online learning problem with the guidance of simple user-provided prompts, thus enabling real-time optimization to satisfy users' personalized preferences. To reduce the computational cost brought by this optimization process for each token, we additionally provide a closed-form solution for each iteration step of the optimization process, thereby reducing the computational time cost to a negligible level. The detailed experimental results demonstrate that Amulet can achieve significant performance improvements in rich settings with combinations of different LLMs, datasets, and user preferences, while maintaining acceptable computational efficiency.