Real-world robot manipulation in dynamic unstructured environments requires lifelong adaptability to evolving objects, scenes and tasks. Traditional imitation learning relies on static training paradigms, which are ill-suited for lifelong adaptation. Although Continual Imitation Learnin (CIL) enables incremental task adaptation while preserving learned knowledge, current CIL methods primarily overlook the intrinsic skill characteristics of robot manipulation or depend on manually defined and rigid skills, leading to suboptimal cross-task knowledge transfer. To address these issues, we propose Skill Prompts-based HiErarchical Continual Imitation Learning (SPECI), a novel end-to-end hierarchical CIL policy architecture for robot manipulation. The SPECI framework consists of a multimodal perception and fusion module for heterogeneous sensory information encoding, a high-level skill inference module for dynamic skill extraction and selection, and a low-level action execution module for precise action generation. To enable efficient knowledge transfer on both skill and task levels, SPECI performs continual implicit skill acquisition and reuse via an expandable skill codebook and an attention-driven skill selection mechanism. Furthermore, we introduce mode approximation to augment the last two modules with task-specific and task-sharing parameters, thereby enhancing task-level knowledge transfer. Extensive experiments on diverse manipulation task suites demonstrate that SPECI consistently outperforms state-of-the-art CIL methods across all evaluated metrics, revealing exceptional bidirectional knowledge transfer and superior overall performance.

Large Language Model (LLM)-based in-application assistants, or copilots, can automate software tasks, but users often prefer learning by doing, raising questions about the optimal level of automation for an effective user experience. We investigated two automation paradigms by designing and implementing a fully automated copilot (AutoCopilot) and a semi-automated copilot (GuidedCopilot) that automates trivial steps while offering step-by-step visual guidance. In a user study (N=20) across data analysis and visual design tasks, GuidedCopilot outperformed AutoCopilot in user control, software utility, and learnability, especially for exploratory and creative tasks, while AutoCopilot saved time for simpler visual tasks. A follow-up design exploration (N=10) enhanced GuidedCopilot with task-and state-aware features, including in-context preview clips and adaptive instructions. Our findings highlight the critical role of user control and tailored guidance in designing the next generation of copilots that enhance productivity, support diverse skill levels, and foster deeper software engagement.

Predominant LLMs focus on high-resource languages while leaving low-resource languages, particularly those in Southeast Asia (SEA), underrepresented. In addition, those models are general-purpose and pay limited attention to the e-commerce domain. To overcome these limitations, we introduce Compass-v2, a lightweight Mixture-of-Experts (MoE) model specifically designed for Southeast Asian languages and e-commerce applications. To balance model performance and inference cost, the model is designed with 30B total parameters and 5B active parameters, incorporating both fine-grained and shared expert modules. To enhance multilingual performance, we curated and constructed a high-quality, industry-leading SEA dataset, to the best of our knowledge. To boost performance in the e-commerce domain, we built a dataset comprising hundreds of billions of tokens, sourced through external data mining and internal platform collection. Besides, we pioneered a hybrid reasoning model that supports both fast thinking and deep thinking within a unified framework to enhance the reasoning capabilities, diverging from the conventional industry practice of deploying two separate models. Through extensive experimental evaluations, our model demonstrates state-of-the-art SEA multilingual and e-commerce performance among sub-30B models, while maintaining significantly lower inference cost.

As large language models (LLMs) continue to advance in linguistic capabilities, robust multilingual evaluation has become essential for promoting equitable technological progress. This position paper examines over 2,000 multilingual (non-English) benchmarks from 148 countries, published between 2021 and 2024, to evaluate past, present, and future practices in multilingual benchmarking. Our findings reveal that, despite significant investments amounting to tens of millions of dollars, English remains significantly overrepresented in these benchmarks. Additionally, most benchmarks rely on original language content rather than translations, with the majority sourced from high-resource countries such as China, India, Germany, the UK, and the USA. Furthermore, a comparison of benchmark performance with human judgments highlights notable disparities. STEM-related tasks exhibit strong correlations with human evaluations (0.70 to 0.85), while traditional NLP tasks like question answering (e.g., XQuAD) show much weaker correlations (0.11 to 0.30). Moreover, translating English benchmarks into other languages proves insufficient, as localized benchmarks demonstrate significantly higher alignment with local human judgments (0.68) than their translated counterparts (0.47). This underscores the importance of creating culturally and linguistically tailored benchmarks rather than relying solely on translations. Through this comprehensive analysis, we highlight six key limitations in current multilingual evaluation practices, propose the guiding principles accordingly for effective multilingual benchmarking, and outline five critical research directions to drive progress in the field. Finally, we call for a global collaborative effort to develop human-aligned benchmarks that prioritize real-world applications.

Recently, Transformer-based robotic manipulation methods utilize multi-view spatial representations and language instructions to learn robot motion trajectories by leveraging numerous robot demonstrations. However, the collection of robot data is extremely challenging, and existing methods lack the capability for continuous learning on new tasks with only a few demonstrations. In this paper, we formulate these challenges as the Few-Shot Action-Incremental Learning (FSAIL) task, and accordingly design a Task-prOmpt graPh evolutIon poliCy (TOPIC) to address these issues. Specifically, to address the data scarcity issue in robotic imitation learning, TOPIC learns Task-Specific Prompts (TSP) through the deep interaction of multi-modal information within few-shot demonstrations, thereby effectively extracting the task-specific discriminative information. On the other hand, to enhance the capability for continual learning on new tasks and mitigate the issue of catastrophic forgetting, TOPIC adopts a Continuous Evolution Strategy (CES). CES leverages the intrinsic relationships between tasks to construct a task relation graph, which effectively facilitates the adaptation of new tasks by reusing skills learned from previous tasks. TOPIC pioneers few-shot continual learning in the robotic manipulation task, and extensive experimental results demonstrate that TOPIC outperforms state-of-the-art baselines by over 26$\%$ in success rate, significantly enhancing the continual learning capabilities of existing Transformer-based policies.

Recent reasoning models through test-time scaling have demonstrated that long chain-of-thoughts can unlock substantial performance boosts in hard reasoning tasks such as math and code. However, the benefit of such long thoughts for system-2 reasoning is relatively less explored in other domains such as perceptual tasks where shallower, system-1 reasoning seems sufficient. In this paper, we introduce LongPerceptualThoughts, a new synthetic dataset with 30K long-thought traces for perceptual tasks. The key challenges in synthesizing elaborate reasoning thoughts for perceptual tasks are that off-the-shelf models are not yet equipped with such thinking behavior and that it is not straightforward to build a reliable process verifier for perceptual tasks. Thus, we propose a novel three-stage data synthesis framework that first synthesizes verifiable multiple-choice questions from dense image descriptions, then extracts simple CoTs from VLMs for those verifiable problems, and finally expands those simple thoughts to elaborate long thoughts via frontier reasoning models. In controlled experiments with a strong instruction-tuned 7B model, we demonstrate notable improvements over existing visual reasoning data-generation methods. Our model, trained on the generated dataset, achieves an average +3.4 points improvement over 5 vision-centric benchmarks, including +11.8 points on V$^*$ Bench. Notably, despite being tuned for vision tasks, it also improves performance on the text reasoning benchmark, MMLU-Pro, by +2 points.

Extractive reading comprehension question answering (QA) datasets are typically evaluated using Exact Match (EM) and F1-score, but these metrics often fail to fully capture model performance. With the success of large language models (LLMs), they have been employed in various tasks, including serving as judges (LLM-as-a-judge). In this paper, we reassess the performance of QA models using LLM-as-a-judge across four reading comprehension QA datasets. We examine different families of LLMs and various answer types to evaluate the effectiveness of LLM-as-a-judge in these tasks. Our results show that LLM-as-a-judge is highly correlated with human judgments and can replace traditional EM/F1 metrics. By using LLM-as-a-judge, the correlation with human judgments improves significantly, from 0.22 (EM) and 0.40 (F1-score) to 0.85. These findings confirm that EM and F1 metrics underestimate the true performance of the QA models. While LLM-as-a-judge is not perfect for more difficult answer types (e.g., job), it still outperforms EM/F1, and we observe no bias issues, such as self-preference, when the same model is used for both the QA and judgment tasks.

As generative artificial intelligence (AI) enables the creation and dissemination of information at massive scale and speed, it is increasingly important to understand how people perceive AI-generated content. One prominent policy proposal requires explicitly labeling AI-generated content to increase transparency and encourage critical thinking about the information, but prior research has not yet tested the effects of such labels. To address this gap, we conducted a survey experiment (N=1601) on a diverse sample of Americans, presenting participants with an AI-generated message about several public policies (e.g., allowing colleges to pay student-athletes), randomly assigning whether participants were told the message was generated by (a) an expert AI model, (b) a human policy expert, or (c) no label. We found that messages were generally persuasive, influencing participants' views of the policies by 9.74 percentage points on average. However, while 94.6% of participants assigned to the AI and human label conditions believed the authorship labels, labels had no significant effects on participants' attitude change toward the policies, judgments of message accuracy, nor intentions to share the message with others. These patterns were robust across a variety of participant characteristics, including prior knowledge of the policy, prior experience with AI, political party, education level, or age. Taken together, these results imply that, while authorship labels would likely enhance transparency, they are unlikely to substantially affect the persuasiveness of the labeled content, highlighting the need for alternative strategies to address challenges posed by AI-generated information.

The 2023 TinyStories study developed an English dataset that allows Small Language Models (SLMs) with 1-10 million parameters to produce coherent outputs matching those of LLMs. Our research expands this framework by creating translated as well as synthetically generated datasets in Indian languages. Using this new dataset, we demonstrate that SLMs efficiently process regional languages with significantly fewer parameters than LLMs, and additionally offer a complementary framework for "inference-based evaluation" of tokenization strategies and linguistic complexity. Our analysis reveals that language-specific tokenizers outperform general-purpose ones for Indian languages. Empirical validations, supported by information-theoretic and morphological analyses, provide insights into the superior performance of Hindi models over Marathi and Bengali. The study uncovers distinct cross-linguistic patterns: Bengali emphasizes creativity, Hindi excels in context understanding and grammar with model scaling, and Marathi requires larger models to capture its unique linguistic features. Optimal parameter allocation varies, with Hindi benefiting more from wider architectures and Bengali favoring a balanced approach. We also show that quality synthetic datasets outperform translated content for training SLMs by 15-30 % . These findings advance both the practical application of SLMs to underserved languages and our theoretical understanding of neural language development.

Federated learning (FL) allows edge devices to collaboratively train models without sharing local data. As FL gains popularity, clients may need to train multiple unrelated FL models, but communication constraints limit their ability to train all models simultaneously. While clients could train FL models sequentially, opportunistically having FL clients concurrently train different models -- termed multi-model federated learning (MMFL) -- can reduce the overall training time. Prior work uses simple client-to-model assignments that do not optimize the contribution of each client to each model over the course of its training. Prior work on single-model FL shows that intelligent client selection can greatly accelerate convergence, but naïve extensions to MMFL can violate heterogeneous resource constraints at both the server and the clients. In this work, we develop a novel convergence analysis of MMFL with arbitrary client sampling methods, theoretically demonstrating the strengths and limitations of previous well-established gradient-based methods. Motivated by this analysis, we propose MMFL-LVR, a loss-based sampling method that minimizes training variance while explicitly respecting communication limits at the server and reducing computational costs at the clients. We extend this to MMFL-StaleVR, which incorporates stale updates for improved efficiency and stability, and MMFL-StaleVRE, a lightweight variant suitable for low-overhead deployment. Experiments show our methods improve average accuracy by up to 19.1% over random sampling, with only a 5.4% gap from the theoretical optimum (full client participation).