Liang Feng

Yan Wang, Wenjie Luo, Junjie Bai et al. · nvidia

End-to-end architectures trained via imitation learning have advanced autonomous driving by scaling model size and data, yet performance remains brittle in safety-critical long-tail scenarios where supervision is sparse and causal understanding is limited. To address this, we introduce Alpamayo-R1 (AR1), a vision-language-action model (VLA) that integrates Chain of Causation reasoning with trajectory planning to enhance decision-making in complex driving scenarios. Our approach features three key innovations: (1) the Chain of Causation (CoC) dataset, built through a hybrid auto-labeling and human-in-the-loop pipeline producing decision-grounded, causally linked reasoning traces aligned with driving behaviors; (2) a modular VLA architecture combining Cosmos-Reason, a Vision-Language Model pre-trained for Physical AI applications, with a diffusion-based trajectory decoder that generates dynamically feasible plans in real time; (3) a multi-stage training strategy using supervised fine-tuning to elicit reasoning and reinforcement learning (RL) to optimize reasoning quality via large reasoning model feedback and enforce reasoning-action consistency. Evaluation shows AR1 achieves up to a 12% improvement in planning accuracy on challenging cases compared to a trajectory-only baseline, with a 35% reduction in off-road rate and 25% reduction in close encounter rate in closed-loop simulation. RL post-training improves reasoning quality by 45% as measured by a large reasoning model critic and reasoning-action consistency by 37%. Model scaling from 0.5B to 7B parameters shows consistent improvements. On-vehicle road tests confirm real-time performance (99 ms latency) and successful urban deployment. By bridging interpretable reasoning with precise control, AR1 demonstrates a practical path towards Level 4 autonomous driving. We plan to release AR1 models and a subset of the CoC in a future update.

Xuebin Lyu, Yuxiao Huang, XueFeng Chen et al.

Evolutionary multi-task optimization (EMTO) is an advanced optimization paradigm that improves search efficiency by enabling knowledge transfer across multiple tasks solved in parallel. Accordingly, a broad range of knowledge transfer methods (KTMs) have been developed as integral components of EMTO algorithms, most of which are tailored to specific problem settings. However, the design of effective KTMs typically relies on substantial domain expertise and careful manual customization, as different EMTO scenarios require distinct transfer strategies to achieve performance gains. Meanwhile, recent advances in large language models (LLMs) have demonstrated strong capabilities in autonomous programming and algorithm synthesis, opening up new possibilities for automating the design of optimization solvers. Motivated by this, in this paper, we propose a Self-guided Knowledge Transfer Design (SKTD) framework that leverages LLMs to autonomously generate knowledge transfer methods (KTMs) as algorithmic components within EMTO. By enabling data-driven and self-adaptive construction of transfer strategies, SKTD facilitates effective knowledge reuse across heterogeneous tasks and diverse EMTO scenarios. To the best of our knowledge, this work represents the first attempt to automate the generation of KTMs for EMTO. Extensive experiments on well-established EMTO benchmarks with varying degrees of task similarity demonstrate that the proposed SKTD consistently achieves superior or highly competitive performance compared with both the state-of-the-art program search approach and manually designed EMTO methods, in terms of optimization effectiveness and cross-scenario generalization.

Yuxiao Huang, Wenjie Zhang, Liang Feng et al.

Recently, large language models (LLMs) have notably positioned them as capable tools for addressing complex optimization challenges. Despite this recognition, a predominant limitation of existing LLM-based optimization methods is their struggle to capture the relationships among decision variables when relying exclusively on numerical text prompts, especially in high-dimensional problems. Keeping this in mind, we first propose to enhance the optimization performance using multimodal LLM capable of processing both textual and visual prompts for deeper insights of the processed optimization problem. This integration allows for a more comprehensive understanding of optimization problems, akin to human cognitive processes. We have developed a multimodal LLM-based optimization framework that simulates human problem-solving workflows, thereby offering a more nuanced and effective analysis. The efficacy of this method is evaluated through extensive empirical studies focused on a well-known combinatorial optimization problem, i.e., capacitated vehicle routing problem. The results are compared against those obtained from the LLM-based optimization algorithms that rely solely on textual prompts, demonstrating the significant advantages of our multimodal approach.

Liang Feng, Qingxia Shang, Yaqing Hou et al.

In recent years, to improve the evolutionary algorithms used to solve optimization problems involving a large number of decision variables, many attempts have been made to simplify the problem solution space of a given problem for the evolutionary search. In the literature, the existing approaches can generally be categorized as decomposition-based methods and dimension-reduction-based methods. The former decomposes a large-scale problem into several smaller subproblems, while the latter transforms the original high-dimensional solution space into a low-dimensional space. However, it is worth noting that a given large-scale optimization problem may not always be decomposable, and it is also difficult to guarantee that the global optimum of the original problem is preserved in the reduced low-dimensional problem space. This paper thus proposes a new search paradigm, namely the multi-space evolutionary search, to enhance the existing evolutionary search methods for solving large-scale optimization problems. In contrast to existing approaches that perform an evolutionary search in a single search space, the proposed paradigm is designed to conduct a search in multiple solution spaces that are derived from the given problem, each possessing a unique landscape. The proposed paradigm makes no assumptions about the large-scale optimization problem of interest, such as that the problem is decomposable or that a certain relationship exists among the decision variables. To verify the efficacy of the proposed paradigm, comprehensive empirical studies in comparison to four state-of-the-art algorithms were conducted using the CEC2013 large-scale benchmark problems.

Zhenzhong Wang, Min Jiang, Xing Gao et al.

Dynamic multi-objective optimization problems (DMOPs) remain a challenge to be settled, because of conflicting objective functions change over time. In recent years, transfer learning has been proven to be a kind of effective approach in solving DMOPs. In this paper, a novel transfer learning based dynamic multi-objective optimization algorithm (DMOA) is proposed called regression transfer learning prediction based DMOA (RTLP-DMOA). The algorithm aims to generate an excellent initial population to accelerate the evolutionary process and improve the evolutionary performance in solving DMOPs. When an environmental change is detected, a regression transfer learning prediction model is constructed by reusing the historical population, which can predict objective values. Then, with the assistance of this prediction model, some high-quality solutions with better predicted objective values are selected as the initial population, which can improve the performance of the evolutionary process. We compare the proposed algorithm with three state-of-the-art algorithms on benchmark functions. Experimental results indicate that the proposed algorithm can significantly enhance the performance of static multi-objective optimization algorithms and is competitive in convergence and diversity.

Yuan Yuan, Yew-Soon Ong, Liang Feng et al.

In this report, we suggest nine test problems for multi-task multi-objective optimization (MTMOO), each of which consists of two multiobjective optimization tasks that need to be solved simultaneously. The relationship between tasks varies between different test problems, which would be helpful to have a comprehensive evaluation of the MO-MFO algorithms. It is expected that the proposed test problems will germinate progress the field of the MTMOO research.

Liang Feng, Yew Soon Ong, Ah Hwee Tan et al.

A significantly under-explored area of evolutionary optimization in the literature is the study of optimization methodologies that can evolve along with the problems solved. Particularly, present evolutionary optimization approaches generally start their search from scratch or the ground-zero state of knowledge, independent of how similar the given new problem of interest is to those optimized previously. There has thus been the apparent lack of automated knowledge transfers and reuse across problems. Taking the cue, this paper introduces a novel Memetic Computational Paradigm for search, one that models after how human solves problems, and embarks on a study towards intelligent evolutionary optimization of problems through the transfers of structured knowledge in the form of memes learned from previous problem-solving experiences, to enhance future evolutionary searches. In particular, the proposed memetic search paradigm is composed of four culture-inspired operators, namely, Meme Learning, Meme Selection, Meme Variation and Meme Imitation. The learning operator mines for memes in the form of latent structures derived from past experiences of problem-solving. The selection operator identifies the fit memes that replicate and transmit across problems, while the variation operator introduces innovations into the memes. The imitation operator, on the other hand, defines how fit memes assimilate into the search process of newly encountered problems, thus gearing towards efficient and effective evolutionary optimization. Finally, comprehensive studies on two widely studied challenging well established NP-hard routing problem domains, particularly, the capacitated vehicle routing (CVR) and capacitated arc routing (CAR), confirm the high efficacy of the proposed memetic computational search paradigm for intelligent evolutionary optimization of problems.