Sadhana Kumaravel

Kinjal Basu, Ibrahim Abdelaziz, Kiran Kate et al. · ibm-research

The resurgence of autonomous agents built using large language models (LLMs) to solve complex real-world tasks has brought increased focus on LLMs' fundamental ability of tool or function calling. At the core of these agents, an LLM must plan, execute, and respond using external tools, APIs, and custom functions. Research on tool calling has gathered momentum, but evaluation benchmarks and datasets representing the complexity of the tasks have lagged behind. In this work, we focus on one such complexity, nested sequencing, with the goal of extending existing benchmarks and evaluation. Specifically, we present NESTFUL, a benchmark to evaluate LLMs on nested sequences of API calls, i.e., sequences where the output of one API call is passed as input to a subsequent call. NESTFUL contains 1800+ nested sequences where all the function calls are executable. Experimental results on a variety of models show that the best-performing model (GPT-4o) achieves a full sequence match accuracy of 28% and a win-rate of 60%, necessitating a large scope for improvement in the nested sequencing aspect of function calling. Our analysis of these results provides possible future research directions for the community, in addition to a benchmark to track progress. We have released the NESTFUL dataset under the Apache 2.0 license at https://github.com/IBM/NESTFUL.

Maxwell Crouse, Ibrahim Abdelaziz, Kshitij Fadnis et al. · ibm-research

Synthetic data has proven itself to be a valuable resource for tuning smaller, cost-effective language models to handle the complexities of multi-turn tool calling conversations. While many frameworks and systems for producing synthetic multi-turn tool calling data have been proposed, prior works have frequently assumed that any tool calling interactions will take place in an execution environment that maintains state. When such an environment is available, this is advantageous as it allows for the validity of an interaction to be determined by whether or not the state of the execution environment matches to some prespecified objective. Unfortunately, this does not hold in many real-world tool use settings, e.g., in enterprise settings where data security is of the utmost importance or in cases where tool specifications are synthesized from multiple sources. In this work, we address this gap by introducing a data generation method, DiGiT-TC, that is designed to produce tool calling conversations that have the characteristics of conversations generated through search in a stateful environment. The key to our technique lies in a novel generation pattern that allows our approach to implicitly represent certain tool calls in the user request. We validate our approach on standard tool calling benchmarks and demonstrate that, even in stateful problem settings, our approach results in strong performance gains.

Maxwell Crouse, Ibrahim Abdelaziz, Ramon Astudillo et al. · ibm-research

Autonomous, goal-driven agents powered by LLMs have recently emerged as promising tools for solving challenging problems without the need for task-specific finetuned models that can be expensive to procure. Currently, the design and implementation of such agents is ad hoc, as the wide variety of tasks that LLM-based agents may be applied to naturally means there can be no one-size-fits-all approach to agent design. In this work we aim to alleviate the difficulty of designing and implementing new agents by proposing a minimalistic generation framework that simplifies the process of building agents. The framework we introduce allows the user to define desired agent behaviors in a high-level, declarative specification that is then used to construct a decoding monitor which guarantees the LLM will produce an output exhibiting the desired behavior. Our declarative approach, in which the behavior is described without concern for how it should be implemented or enforced, enables rapid design, implementation, and experimentation with different LLM-based agents. We demonstrate how the proposed framework can be used to implement recent LLM-based agents (e.g., ReACT), and show how the flexibility of our approach can be leveraged to define a new agent with more complex behavior, the Plan-Act-Summarize-Solve (PASS) agent. Lastly, we demonstrate that our method outperforms other agents on multiple popular reasoning-centric question-answering benchmarks.

Kinjal Basu, Ibrahim Abdelaziz, Subhajit Chaudhury et al. · ibm-research

There is a growing need for Large Language Models (LLMs) to effectively use tools and external Application Programming Interfaces (APIs) to plan and complete tasks. As such, there is tremendous interest in methods that can acquire sufficient quantities of train and test data that involve calls to tools / APIs. Two lines of research have emerged as the predominant strategies for addressing this challenge. The first has focused on synthetic data generation techniques, while the second has involved curating task-adjacent datasets which can be transformed into API / Tool-based tasks. In this paper, we focus on the task of identifying, curating, and transforming existing datasets and, in turn, introduce API-BLEND, a large corpora for training and systematic testing of tool-augmented LLMs. The datasets mimic real-world scenarios involving API-tasks such as API / tool detection, slot filling, and sequencing of the detected APIs. We demonstrate the utility of the API-BLEND dataset for both training and benchmarking purposes.

Ibrahim Abdelaziz, Kinjal Basu, Mayank Agarwal et al.

Large language models (LLMs) have recently shown tremendous promise in serving as the backbone to agentic systems, as demonstrated by their performance in multi-faceted, challenging benchmarks like SWE-Bench and Agent-Bench. However, to realize the true potential of LLMs as autonomous agents, they must learn to identify, call, and interact with external tools and application program interfaces (APIs) to complete complex tasks. These tasks together are termed function calling. Endowing LLMs with function calling abilities leads to a myriad of advantages, such as access to current and domain-specific information in databases and knowledge sources, and the ability to outsource tasks that can be reliably performed by tools, e.g., a Python interpreter or calculator. While there has been significant progress in function calling with LLMs, there is still a dearth of open models that perform on par with proprietary LLMs like GPT, Claude, and Gemini. Therefore, in this work, we introduce the GRANITE-20B-FUNCTIONCALLING model under an Apache 2.0 license. The model is trained using a multi-task training approach on seven fundamental tasks encompassed in function calling, those being Nested Function Calling, Function Chaining, Parallel Functions, Function Name Detection, Parameter-Value Pair Detection, Next-Best Function, and Response Generation. We present a comprehensive evaluation on multiple out-of-domain datasets comparing GRANITE-20B-FUNCTIONCALLING to more than 15 other best proprietary and open models. GRANITE-20B-FUNCTIONCALLING provides the best performance among all open models on the Berkeley Function Calling Leaderboard and fourth overall. As a result of the diverse tasks and datasets used for training our model, we show that GRANITE-20B-FUNCTIONCALLING has better generalizability on multiple tasks in seven different evaluation datasets.

Sadhana Kumaravel, Tahira Naseem, Ramon Fernandez Astudillo et al.

The sliding window approach provides an elegant way to handle contexts of sizes larger than the Transformer's input window, for tasks like language modeling. Here we extend this approach to the sequence-to-sequence task of document parsing. For this, we exploit recent progress in transition-based parsing to implement a parser with synchronous sliding windows over source and target. We develop an oracle and a parser for document-level AMR by expanding on Structured-BART such that it leverages source-target alignments and constrains decoding to guarantee synchronicity and consistency across overlapping windows. We evaluate our oracle and parser using the Abstract Meaning Representation (AMR) parsing 3.0 corpus. On the Multi-Sentence development set of AMR 3.0, we show that our transition oracle loses only 8\% of the gold cross-sentential links despite using a sliding window. In practice, this approach also results in a high-quality document-level parser with manageable memory requirements. Our proposed system performs on par with the state-of-the-art pipeline approach for document-level AMR parsing task on Multi-Sentence AMR 3.0 corpus while maintaining sentence-level parsing performance.

Tahira Naseem, Austin Blodgett, Sadhana Kumaravel et al.

Despite extensive research on parsing of English sentences into Abstraction Meaning Representation (AMR) graphs, which are compared to gold graphs via the Smatch metric, full-document parsing into a unified graph representation lacks well-defined representation and evaluation. Taking advantage of a super-sentential level of coreference annotation from previous work, we introduce a simple algorithm for deriving a unified graph representation, avoiding the pitfalls of information loss from over-merging and lack of coherence from under-merging. Next, we describe improvements to the Smatch metric to make it tractable for comparing document-level graphs, and use it to re-evaluate the best published document-level AMR parser. We also present a pipeline approach combining the top performing AMR parser and coreference resolution systems, providing a strong baseline for future research.

Wei Zhang, Murray Campbell, Yang Yu et al.

Human judgments of word similarity have been a popular method of evaluating the quality of word embedding. But it fails to measure the geometry properties such as asymmetry. For example, it is more natural to say "Ellipses are like Circles" than "Circles are like Ellipses". Such asymmetry has been observed from a psychoanalysis test called word evocation experiment, where one word is used to recall another. Although useful, such experimental data have been significantly understudied for measuring embedding quality. In this paper, we use three well-known evocation datasets to gain insights into asymmetry encoding of embedding. We study both static embedding as well as contextual embedding, such as BERT. Evaluating asymmetry for BERT is generally hard due to the dynamic nature of embedding. Thus, we probe BERT's conditional probabilities (as a language model) using a large number of Wikipedia contexts to derive a theoretically justifiable Bayesian asymmetry score. The result shows that contextual embedding shows randomness than static embedding on similarity judgments while performing well on asymmetry judgment, which aligns with its strong performance on "extrinsic evaluations" such as text classification. The asymmetry judgment and the Bayesian approach provides a new perspective to evaluate contextual embedding on intrinsic evaluation, and its comparison to similarity evaluation concludes our work with a discussion on the current state and the future of representation learning.

Keerthiram Murugesan, Mattia Atzeni, Pavan Kapanipathi et al.

Text-based games have emerged as an important test-bed for Reinforcement Learning (RL) research, requiring RL agents to combine grounded language understanding with sequential decision making. In this paper, we examine the problem of infusing RL agents with commonsense knowledge. Such knowledge would allow agents to efficiently act in the world by pruning out implausible actions, and to perform look-ahead planning to determine how current actions might affect future world states. We design a new text-based gaming environment called TextWorld Commonsense (TWC) for training and evaluating RL agents with a specific kind of commonsense knowledge about objects, their attributes, and affordances. We also introduce several baseline RL agents which track the sequential context and dynamically retrieve the relevant commonsense knowledge from ConceptNet. We show that agents which incorporate commonsense knowledge in TWC perform better, while acting more efficiently. We conduct user-studies to estimate human performance on TWC and show that there is ample room for future improvement.

Anna Choromanska, Benjamin Cowen, Sadhana Kumaravel et al.

Despite significant recent advances in deep neural networks, training them remains a challenge due to the highly non-convex nature of the objective function. State-of-the-art methods rely on error backpropagation, which suffers from several well-known issues, such as vanishing and exploding gradients, inability to handle non-differentiable nonlinearities and to parallelize weight-updates across layers, and biological implausibility. These limitations continue to motivate exploration of alternative training algorithms, including several recently proposed auxiliary-variable methods which break the complex nested objective function into local subproblems. However, those techniques are mainly offline (batch), which limits their applicability to extremely large datasets, as well as to online, continual or reinforcement learning. The main contribution of our work is a novel online (stochastic/mini-batch) alternating minimization (AM) approach for training deep neural networks, together with the first theoretical convergence guarantees for AM in stochastic settings and promising empirical results on a variety of architectures and datasets.