ToolGen: Unified Tool Retrieval and Calling via Generation

As large language models (LLMs) advance, their inability to autonomously execute tasks by directly interacting with external tools remains a critical limitation. Traditional methods rely on inputting tool descriptions as context, which is constrained by context length and requires separate, often inefficient, retrieval mechanisms. We introduce ToolGen, a paradigm shift that integrates tool knowledge directly into the LLM's parameters by representing each tool as a unique token. This enables the LLM to generate tool calls and arguments as part of its next token prediction capabilities, seamlessly blending tool invocation with language generation. Our framework allows the LLM to access and utilize a vast amount of tools with no additional retrieval step, significantly enhancing both performance and scalability. Experimental results with over 47,000 tools show that ToolGen not only achieves superior results in both tool retrieval and autonomous task completion but also sets the stage for a new era of AI agents that can adapt to tools across diverse domains. By fundamentally transforming tool retrieval into a generative process, ToolGen paves the way for more versatile, efficient, and autonomous AI systems. ToolGen enables end-to-end tool learning and opens opportunities for integration with other advanced techniques such as chain-of-thought and reinforcement learning, thereby expanding the practical capabilities of LLMs.

SimBench: A Rule-Based Multi-Turn Interaction Benchmark for Evaluating an LLM's Ability to Generate Digital Twins

We introduce SimBench, a benchmark designed to evaluate the proficiency of student large language models (S-LLMs) in generating digital twins (DTs) that can be used in simulators for virtual testing. Given a collection of S-LLMs, this benchmark enables the ranking of the S-LLMs based on their ability to produce high-quality DTs. We demonstrate this by comparing over 20 open- and closed-source S-LLMs. Using multi-turn interactions, SimBench employs a rule-based judge LLM (J-LLM) that leverages both predefined rules and human-in-the-loop guidance to assign scores for the DTs generated by the S-LLM, thus providing a consistent and expert-inspired evaluation protocol. The J-LLM is specific to a simulator, and herein the proposed benchmarking approach is demonstrated in conjunction with the Chrono multi-physics simulator. Chrono provided the backdrop used to assess an S-LLM in relation to the latter's ability to create digital twins for multibody dynamics, finite element analysis, vehicle dynamics, robotic dynamics, and sensor simulations. The proposed benchmarking principle is broadly applicable and enables the assessment of an S-LLM's ability to generate digital twins for other simulation packages. All code and data are available at https://github.com/uwsbel/SimBench.

Raising the Bar: Investigating the Values of Large Language Models via Generative Evolving Testing

Warning: this paper contains model outputs exhibiting unethical information. Large Language Models (LLMs) have achieved significant breakthroughs, but their generated unethical content poses potential risks. Measuring value alignment of LLMs becomes crucial for their regulation and responsible deployment. Numerous datasets have been constructed to assess social bias, toxicity, and ethics in LLMs, but they suffer from evaluation chronoeffect, that is, as models rapidly evolve, existing data becomes leaked or undemanding, overestimating ever-developing LLMs. To tackle this problem, we propose GETA, a novel generative evolving testing approach that dynamically probes the underlying moral baselines of LLMs. Distinct from previous adaptive testing methods that rely on static datasets with limited difficulty, GETA incorporates an iteratively-updated item generator which infers each LLM's moral boundaries and generates difficulty-tailored testing items, accurately reflecting the true alignment extent. This process theoretically learns a joint distribution of item and model response, with item difficulty and value conformity as latent variables, where the generator co-evolves with the LLM, addressing chronoeffect. We evaluate various popular LLMs with diverse capabilities and demonstrate that GETA can create difficulty-matching testing items and more accurately assess LLMs' values, better consistent with their performance on unseen OOD and i.i.d. items, laying the groundwork for future evaluation paradigms.

The Surprising Benefits of Base Rate Neglect in Robust Aggregation

Robust aggregation integrates predictions from multiple experts without knowledge of the experts' information structures. Prior work assumes experts are Bayesian, providing predictions as perfect posteriors based on their signals. However, real-world experts often deviate systematically from Bayesian reasoning. Our work considers experts who tend to ignore the base rate. We find that a certain degree of base rate neglect helps with robust forecast aggregation. Specifically, we consider a forecast aggregation problem with two experts who each predict a binary world state after observing private signals. Unlike previous work, we model experts exhibiting base rate neglect, where they incorporate the base rate information to degree $\lambda\in[0,1]$, with $\lambda=0$ indicating complete ignorance and $\lambda=1$ perfect Bayesian updating. To evaluate aggregators' performance, we adopt Arieli et al. (2018)'s worst-case regret model, which measures the maximum regret across the set of considered information structures compared to an omniscient benchmark. Our results reveal the surprising V-shape of regret as a function of $\lambda$. That is, predictions with an intermediate incorporating degree of base rate $\lambda<1$ can counter-intuitively lead to lower regret than perfect Bayesian posteriors with $\lambda=1$. We additionally propose a new aggregator with low regret robust to unknown $\lambda$. Finally, we conduct an empirical study to test the base rate neglect model and evaluate the performance of various aggregators.

Ag2Manip: Learning Novel Manipulation Skills with Agent-Agnostic Visual and Action Representations

Autonomous robotic systems capable of learning novel manipulation tasks are poised to transform industries from manufacturing to service automation. However, modern methods (e.g., VIP and R3M) still face significant hurdles, notably the domain gap among robotic embodiments and the sparsity of successful task executions within specific action spaces, resulting in misaligned and ambiguous task representations. We introduce Ag2Manip (Agent-Agnostic representations for Manipulation), a framework aimed at surmounting these challenges through two key innovations: a novel agent-agnostic visual representation derived from human manipulation videos, with the specifics of embodiments obscured to enhance generalizability; and an agent-agnostic action representation abstracting a robot's kinematics to a universal agent proxy, emphasizing crucial interactions between end-effector and object. Ag2Manip's empirical validation across simulated benchmarks like FrankaKitchen, ManiSkill, and PartManip shows a 325% increase in performance, achieved without domain-specific demonstrations. Ablation studies underline the essential contributions of the visual and action representations to this success. Extending our evaluations to the real world, Ag2Manip significantly improves imitation learning success rates from 50% to 77.5%, demonstrating its effectiveness and generalizability across both simulated and physical environments.

LLM3:Large Language Model-based Task and Motion Planning with Motion Failure Reasoning

Conventional Task and Motion Planning (TAMP) approaches rely on manually crafted interfaces connecting symbolic task planning with continuous motion generation. These domain-specific and labor-intensive modules are limited in addressing emerging tasks in real-world settings. Here, we present LLM^3, a novel Large Language Model (LLM)-based TAMP framework featuring a domain-independent interface. Specifically, we leverage the powerful reasoning and planning capabilities of pre-trained LLMs to propose symbolic action sequences and select continuous action parameters for motion planning. Crucially, LLM^3 incorporates motion planning feedback through prompting, allowing the LLM to iteratively refine its proposals by reasoning about motion failure. Consequently, LLM^3 interfaces between task planning and motion planning, alleviating the intricate design process of handling domain-specific messages between them. Through a series of simulations in a box-packing domain, we quantitatively demonstrate the effectiveness of LLM^3 in solving TAMP problems and the efficiency in selecting action parameters. Ablation studies underscore the significant contribution of motion failure reasoning to the success of LLM^3. Furthermore, we conduct qualitative experiments on a physical manipulator, demonstrating the practical applicability of our approach in real-world settings.

GeoGalactica: A Scientific Large Language Model in Geoscience

Large language models (LLMs) have achieved huge success for their general knowledge and ability to solve a wide spectrum of tasks in natural language processing (NLP). Due to their impressive abilities, LLMs have shed light on potential inter-discipline applications to foster scientific discoveries of a specific domain by using artificial intelligence (AI for science, AI4S). In the meantime, utilizing NLP techniques in geoscience research and practice is wide and convoluted, contributing from knowledge extraction and document classification to question answering and knowledge discovery. In this work, we take the initial step to leverage LLM for science, through a rather straightforward approach. We try to specialize an LLM into geoscience, by further pre-training the model with a vast amount of texts in geoscience, as well as supervised fine-tuning (SFT) the resulting model with our custom collected instruction tuning dataset. These efforts result in a model GeoGalactica consisting of 30 billion parameters. To our best knowledge, it is the largest language model for the geoscience domain. More specifically, GeoGalactica is from further pre-training of Galactica. We train GeoGalactica over a geoscience-related text corpus containing 65 billion tokens curated from extensive data sources in the big science project Deep-time Digital Earth (DDE), preserving as the largest geoscience-specific text corpus. Then we fine-tune the model with 1 million pairs of instruction-tuning data consisting of questions that demand professional geoscience knowledge to answer. In this technical report, we will illustrate in detail all aspects of GeoGalactica, including data collection, data cleaning, base model selection, pre-training, SFT, and evaluation. We open-source our data curation tools and the checkpoints of GeoGalactica during the first 3/4 of pre-training.

A Multimodal Ecological Civilization Pattern Recommendation Method Based on Large Language Models and Knowledge Graph

The Ecological Civilization Pattern Recommendation System (ECPRS) aims to recommend suitable ecological civilization patterns for target regions, promoting sustainable development and reducing regional disparities. However, the current representative recommendation methods are not suitable for recommending ecological civilization patterns in a geographical context. There are two reasons for this. Firstly, regions have spatial heterogeneity, and the (ECPRS)needs to consider factors like climate, topography, vegetation, etc., to recommend civilization patterns adapted to specific ecological environments, ensuring the feasibility and practicality of the recommendations. Secondly, the abstract features of the ecological civilization patterns in the real world have not been fully utilized., resulting in poor richness in their embedding representations and consequently, lower performance of the recommendation system. Considering these limitations, we propose the ECPR-MML method. Initially, based on the novel method UGPIG, we construct a knowledge graph to extract regional representations incorporating spatial heterogeneity features. Following that, inspired by the significant progress made by Large Language Models (LLMs) in the field of Natural Language Processing (NLP), we employ Large LLMs to generate multimodal features for ecological civilization patterns in the form of text and images. We extract and integrate these multimodal features to obtain semantically rich representations of ecological civilization. Through extensive experiments, we validate the performance of our ECPR-MML model. Our results show that F1@5 is 2.11% higher compared to state-of-the-art models, 2.02% higher than NGCF, and 1.16% higher than UGPIG. Furthermore, multimodal data can indeed enhance recommendation performance. However, the data generated by LLM is not as effective as real data to a certain extent.

Hybrid Reinforcement Learning for Optimizing Pump Sustainability in Real-World Water Distribution Networks

This article addresses the pump-scheduling optimization problem to enhance real-time control of real-world water distribution networks (WDNs). Our primary objectives are to adhere to physical operational constraints while reducing energy consumption and operational costs. Traditional optimization techniques, such as evolution-based and genetic algorithms, often fall short due to their lack of convergence guarantees. Conversely, reinforcement learning (RL) stands out for its adaptability to uncertainties and reduced inference time, enabling real-time responsiveness. However, the effective implementation of RL is contingent on building accurate simulation models for WDNs, and prior applications have been limited by errors in simulation training data. These errors can potentially cause the RL agent to learn misleading patterns and actions and recommend suboptimal operational strategies. To overcome these challenges, we present an improved "hybrid RL" methodology. This method integrates the benefits of RL while anchoring it in historical data, which serves as a baseline to incrementally introduce optimal control recommendations. By leveraging operational data as a foundation for the agent's actions, we enhance the explainability of the agent's actions, foster more robust recommendations, and minimize error. Our findings demonstrate that the hybrid RL agent can significantly improve sustainability, operational efficiency, and dynamically adapt to emerging scenarios in real-world WDNs.

Unveiling Optimal SDG Pathways: An Innovative Approach Leveraging Graph Pruning and Intent Graph for Effective Recommendations

The recommendation of appropriate development pathways, also known as ecological civilization patterns for achieving Sustainable Development Goals (namely, sustainable development patterns), are of utmost importance for promoting ecological, economic, social, and resource sustainability in a specific region. To achieve this, the recommendation process must carefully consider the region's natural, environmental, resource, and economic characteristics. However, current recommendation algorithms in the field of computer science fall short in adequately addressing the spatial heterogeneity related to environment and sparsity of regional historical interaction data, which limits their effectiveness in recommending sustainable development patterns. To overcome these challenges, this paper proposes a method called User Graph after Pruning and Intent Graph (UGPIG). Firstly, we utilize the high-density linking capability of the pruned User Graph to address the issue of spatial heterogeneity neglect in recommendation algorithms. Secondly, we construct an Intent Graph by incorporating the intent network, which captures the preferences for attributes including environmental elements of target regions. This approach effectively alleviates the problem of sparse historical interaction data in the region. Through extensive experiments, we demonstrate that UGPIG outperforms state-of-the-art recommendation algorithms like KGCN, KGAT, and KGIN in sustainable development pattern recommendations, with a maximum improvement of 9.61% in Top-3 recommendation performance.