Pavan Kapanipathi

Ibrahim Abdelaziz, Asim Munawar, Kinjal Basu et al.

Training LLMs to orchestrate multi-step tool calls is held back by three coupled obstacles: realistic stateful execution environments are costly to build, synthetic training queries are often detached from the server's actual state (so the generated tool calls fail to execute), and recall-based RL rewards incentivize verbose tool-calling patterns. We present PROVE (Programmatic Rewards On Verified Environments), a framework with three contributions: (1) a library of 20 stateful MCP (Model Context Protocol) servers exposing 343 tools, enabling live-execution RL training with session-scoped state isolation; (2) a state-machine data synthesis pipeline that generates multi-turn tool-call trajectories grounded in live-sampled server state, so generated queries reference entities that actually exist; and (3) a multi-component programmatic reward with an adaptive efficiency penalty that counters the verbosity incentive of recall-based rewards. We train four models (Qwen3-4B, Qwen3-8B, Qwen2.5-7B, Granite-4.1-8B) with GRPO on the resulting ~13K training examples. On BFCL Multi-Turn, tau2-bench, and T-Eval, PROVE yields improvements of up to +10.2, +6.8, and +6.5 points respectively, demonstrating that this framework yields consistent gains on multi-step tool orchestration across two model families.

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.

Dung Thai, Srinivas Ravishankar, Ibrahim Abdelaziz et al.

Knowledge bases (KBs) are often incomplete and constantly changing in practice. Yet, in many question answering applications coupled with knowledge bases, the sparse nature of KBs is often overlooked. To this end, we propose a case-based reasoning approach, CBR-iKB, for knowledge base question answering (KBQA) with incomplete-KB as our main focus. Our method ensembles decisions from multiple reasoning chains with a novel nonparametric reasoning algorithm. By design, CBR-iKB can seamlessly adapt to changes in KBs without any task-specific training or fine-tuning. Our method achieves 100% accuracy on MetaQA and establishes new state-of-the-art on multiple benchmarks. For instance, CBR-iKB achieves an accuracy of 70% on WebQSP under the incomplete-KB setting, outperforming the existing state-of-the-art method by 22.3%.

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.

Subhajit Chaudhury, Sarathkrishna Swaminathan, Daiki Kimura et al.

Text-based reinforcement learning agents have predominantly been neural network-based models with embeddings-based representation, learning uninterpretable policies that often do not generalize well to unseen games. On the other hand, neuro-symbolic methods, specifically those that leverage an intermediate formal representation, are gaining significant attention in language understanding tasks. This is because of their advantages ranging from inherent interpretability, the lesser requirement of training data, and being generalizable in scenarios with unseen data. Therefore, in this paper, we propose a modular, NEuro-Symbolic Textual Agent (NESTA) that combines a generic semantic parser with a rule induction system to learn abstract interpretable rules as policies. Our experiments on established text-based game benchmarks show that the proposed NESTA method outperforms deep reinforcement learning-based techniques by achieving better generalization to unseen test games and learning from fewer training interactions.

Keerthiram Murugesan, Sarathkrishna Swaminathan, Soham Dan et al.

With the growing interest in large language models, the need for evaluating the quality of machine text compared to reference (typically human-generated) text has become focal attention. Most recent works focus either on task-specific evaluation metrics or study the properties of machine-generated text captured by the existing metrics. In this work, we propose a new evaluation scheme to model human judgments in 7 NLP tasks, based on the fine-grained mismatches between a pair of texts. Inspired by the recent efforts in several NLP tasks for fine-grained evaluation, we introduce a set of 13 mismatch error types such as spatial/geographic errors, entity errors, etc, to guide the model for better prediction of human judgments. We propose a neural framework for evaluating machine texts that uses these mismatch error types as auxiliary tasks and re-purposes the existing single-number evaluation metrics as additional scalar features, in addition to textual features extracted from the machine and reference texts. Our experiments reveal key insights about the existing metrics via the mismatch errors. We show that the mismatch errors between the sentence pairs on the held-out datasets from 7 NLP tasks align well with the human evaluation.

Mayank Mishra, Matt Stallone, Gaoyuan Zhang et al. · ibm-research

Large Language Models (LLMs) trained on code are revolutionizing the software development process. Increasingly, code LLMs are being integrated into software development environments to improve the productivity of human programmers, and LLM-based agents are beginning to show promise for handling complex tasks autonomously. Realizing the full potential of code LLMs requires a wide range of capabilities, including code generation, fixing bugs, explaining and documenting code, maintaining repositories, and more. In this work, we introduce the Granite series of decoder-only code models for code generative tasks, trained with code written in 116 programming languages. The Granite Code models family consists of models ranging in size from 3 to 34 billion parameters, suitable for applications ranging from complex application modernization tasks to on-device memory-constrained use cases. Evaluation on a comprehensive set of tasks demonstrates that Granite Code models consistently reaches state-of-the-art performance among available open-source code LLMs. The Granite Code model family was optimized for enterprise software development workflows and performs well across a range of coding tasks (e.g. code generation, fixing and explanation), making it a versatile all around code model. We release all our Granite Code models under an Apache 2.0 license for both research and commercial use.

Sumit Neelam, Udit Sharma, Hima Karanam et al.

Knowledge Base Question Answering (KBQA) tasks that involve complex reasoning are emerging as an important research direction. However, most existing KBQA datasets focus primarily on generic multi-hop reasoning over explicit facts, largely ignoring other reasoning types such as temporal, spatial, and taxonomic reasoning. In this paper, we present a benchmark dataset for temporal reasoning, TempQA-WD, to encourage research in extending the present approaches to target a more challenging set of complex reasoning tasks. Specifically, our benchmark is a temporal question answering dataset with the following advantages: (a) it is based on Wikidata, which is the most frequently curated, openly available knowledge base, (b) it includes intermediate sparql queries to facilitate the evaluation of semantic parsing based approaches for KBQA, and (c) it generalizes to multiple knowledge bases: Freebase and Wikidata. The TempQA-WD dataset is available at https://github.com/IBM/tempqa-wd.

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.

Mingjian Jiang, Yangjun Ruan, Luis Lastras et al. · utoronto

Recent advances in language model (LM) agents have demonstrated significant potential for automating complex real-world tasks. To make progress on these difficult tasks, LM agent architectures have become increasingly complex, often incorporating multi-step retrieval tools, multiple agents, and scaffolding adapted to the underlying LM. In this work, we investigate whether all of this complexity is necessary, or if parts of these scaffolds can be removed on challenging tasks like SWE-bench. We show that in the case of SWE-bench, simply putting the entire environment into the context of a long context language model (LCLM) and properly prompting the model makes it competitive with carefully tuned, complex agent scaffolds. We show that a Gemini-1.5-Pro model without any scaffolding or tools achieves 38% on SWE-Bench-Verified, comparable with approaches using carefully tuned agent scaffolds (32%). While the unscaffolded approach with Gemini-1.5-Pro falls short of the strongest agentic architectures, we demonstrate that the more capable Gemini-2.5-Pro using the same unscaffolded approach directly attains a 50.8% solve rate. Additionally, a two-stage approach combining Gemini-1.5-Pro with Claude-3.7 achieves a competitive 48.6% solve rate.

Tenghao Huang, Kinjal Basu, Ibrahim Abdelaziz et al.

The proliferation of web agents necessitates advanced navigation and interaction strategies within complex web environments. Current models often struggle with efficient navigation and action execution due to limited visibility and understanding of web structures. Our proposed R2D2 framework addresses these challenges by integrating two paradigms: Remember and Reflect. The Remember paradigm uses a replay buffer that aids agents in reconstructing the web environment dynamically, thus enabling the formulation of a detailed "map" of previously visited pages. This helps in reducing navigational errors and optimizing the decision-making process during web interactions. Conversely, the Reflect paradigm allows agents to learn from past mistakes by providing a mechanism for error analysis and strategy refinement, enhancing overall task performance. We evaluate R2D2 using the WebArena benchmark, demonstrating substantial improvements over existing methods, including a 50% reduction in navigation errors and a threefold increase in task completion rates. Our findings suggest that a combination of memory-enhanced navigation and reflective learning promisingly advances the capabilities of web agents, potentially benefiting various applications such as automated customer service and personal digital assistants.

Mayank Agarwal, Ibrahim Abdelaziz, Kinjal Basu et al.

As large language models (LLMs) increasingly interact with external tools, reward modeling for tool use has become a critical yet underexplored area. Existing reward models, trained primarily on natural language outputs, struggle to evaluate tool-based reasoning and execution. To quantify this gap, we introduce FC-RewardBench, the first benchmark designed to systematically assess reward models' performance in tool-calling scenarios. Our analysis shows that current reward models often miss key signals of effective tool use, highlighting the need for domain-specific modeling. To address this, we propose a training framework for outcome-based reward models using data synthesized from permissively licensed, open-weight LLMs. We train models ranging from 1.7B to 14B parameters and evaluate them across seven out-of-domain benchmarks. These models consistently outperform general-purpose baselines, achieving up to 25\% average improvement in downstream task performance and enabling data-efficient fine-tuning through reward-guided filtering.

Peter Phan, Dhruv Agarwal, Kavitha Srinivas et al. · ibm-research

Large language models (LLMs) are increasingly being applied to black-box optimization tasks, from program synthesis to molecule design. Prior work typically leverages in-context learning to iteratively guide the model towards better solutions. Such methods, however, often struggle to balance exploration of new solution spaces with exploitation of high-reward ones. Recently, test-time training (TTT) with synthetic data has shown promise in improving solution quality. However, the need for hand-crafted training data tailored to each task limits feasibility and scalability across domains. To address this problem, we introduce MiGrATe-a method for online TTT that uses GRPO as a search algorithm to adapt LLMs at inference without requiring external training data. MiGrATe operates via a mixed-policy group construction procedure that combines on-policy sampling with two off-policy data selection techniques: greedy sampling, which selects top-performing past completions, and neighborhood sampling (NS), which generates completions structurally similar to high-reward ones. Together, these components bias the policy gradient towards exploitation of promising regions in solution space, while preserving exploration through on-policy sampling. We evaluate MiGrATe on three challenging domains-word search, molecule optimization, and hypothesis+program induction on the Abstraction and Reasoning Corpus (ARC)-and find that it consistently outperforms both inference-only and TTT baselines, demonstrating the potential of online TTT as a solution for complex search tasks without external supervision.

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.

Maxwell Crouse, Ramon Astudillo, Tahira Naseem et al.

We introduce Logical Offline Cycle Consistency Optimization (LOCCO), a scalable, semi-supervised method for training a neural semantic parser. Conceptually, LOCCO can be viewed as a form of self-learning where the semantic parser being trained is used to generate annotations for unlabeled text that are then used as new supervision. To increase the quality of annotations, our method utilizes a count-based prior over valid formal meaning representations and a cycle-consistency score produced by a neural text generation model as additional signals. Both the prior and semantic parser are updated in an alternate fashion from full passes over the training data, which can be seen as approximating the marginalization of latent structures through stochastic variational inference. The use of a count-based prior, frozen text generation model, and offline annotation process yields an approach with negligible complexity and latency increases as compared to conventional self-learning. As an added bonus, the annotations produced by LOCCO can be trivially repurposed to train a neural text generation model. We demonstrate the utility of LOCCO on the well-known WebNLG benchmark where we obtain an improvement of 2 points against a self-learning parser under equivalent conditions, an improvement of 1.3 points against the previous state-of-the-art parser, and competitive text generation performance in terms of BLEU score.

Maxwell Crouse, Pavan Kapanipathi, Subhajit Chaudhury et al.

Nearly all general-purpose neural semantic parsers generate logical forms in a strictly top-down autoregressive fashion. Though such systems have achieved impressive results across a variety of datasets and domains, recent works have called into question whether they are ultimately limited in their ability to compositionally generalize. In this work, we approach semantic parsing from, quite literally, the opposite direction; that is, we introduce a neural semantic parsing generation method that constructs logical forms from the bottom up, beginning from the logical form's leaves. The system we introduce is lazy in that it incrementally builds up a set of potential semantic parses, but only expands and processes the most promising candidate parses at each generation step. Such a parsimonious expansion scheme allows the system to maintain an arbitrarily large set of parse hypotheses that are never realized and thus incur minimal computational overhead. We evaluate our approach on compositional generalization; specifically, on the challenging CFQ dataset and three Text-to-SQL datasets where we show that our novel, bottom-up semantic parsing technique outperforms general-purpose semantic parsers while also being competitive with comparable neural parsers that have been designed for each task.

Mihaela Bornea, Ramon Fernandez Astudillo, Tahira Naseem et al.

We propose a transition-based system to transpile Abstract Meaning Representation (AMR) into SPARQL for Knowledge Base Question Answering (KBQA). This allows us to delegate part of the semantic representation to a strongly pre-trained semantic parser, while learning transpiling with small amount of paired data. We depart from recent work relating AMR and SPARQL constructs, but rather than applying a set of rules, we teach a BART model to selectively use these relations. Further, we avoid explicitly encoding AMR but rather encode the parser state in the attention mechanism of BART, following recent semantic parsing works. The resulting model is simple, provides supporting text for its decisions, and outperforms recent approaches in KBQA across two knowledge bases: DBPedia (LC-QuAD 1.0, QALD-9) and Wikidata (WebQSP, SWQ-WD).

Srinivas Ravishankar, June Thai, Ibrahim Abdelaziz et al.

Most existing approaches for Knowledge Base Question Answering (KBQA) focus on a specific underlying knowledge base either because of inherent assumptions in the approach, or because evaluating it on a different knowledge base requires non-trivial changes. However, many popular knowledge bases share similarities in their underlying schemas that can be leveraged to facilitate generalization across knowledge bases. To achieve this generalization, we introduce a KBQA framework based on a 2-stage architecture that explicitly separates semantic parsing from the knowledge base interaction, facilitating transfer learning across datasets and knowledge graphs. We show that pretraining on datasets with a different underlying knowledge base can nevertheless provide significant performance gains and reduce sample complexity. Our approach achieves comparable or state-of-the-art performance for LC-QuAD (DBpedia), WebQSP (Freebase), SimpleQuestions (Wikidata) and MetaQA (Wikimovies-KG).

Sumit Neelam, Udit Sharma, Hima Karanam et al.

Knowledge Base Question Answering (KBQA) tasks that in-volve complex reasoning are emerging as an important re-search direction. However, most KBQA systems struggle withgeneralizability, particularly on two dimensions: (a) acrossmultiple reasoning types where both datasets and systems haveprimarily focused on multi-hop reasoning, and (b) across mul-tiple knowledge bases, where KBQA approaches are specif-ically tuned to a single knowledge base. In this paper, wepresent SYGMA, a modular approach facilitating general-izability across multiple knowledge bases and multiple rea-soning types. Specifically, SYGMA contains three high levelmodules: 1) KB-agnostic question understanding module thatis common across KBs 2) Rules to support additional reason-ing types and 3) KB-specific question mapping and answeringmodule to address the KB-specific aspects of the answer ex-traction. We demonstrate effectiveness of our system by evalu-ating on datasets belonging to two distinct knowledge bases,DBpedia and Wikidata. In addition, to demonstrate extensi-bility to additional reasoning types we evaluate on multi-hopreasoning datasets and a new Temporal KBQA benchmarkdataset on Wikidata, namedTempQA-WD1, introduced in thispaper. We show that our generalizable approach has bettercompetetive performance on multiple datasets on DBpediaand Wikidata that requires both multi-hop and temporal rea-soning

Prithviraj Sen, Breno W. S. R. Carvalho, Ibrahim Abdelaziz et al.

Recent interest in Knowledge Base Completion (KBC) has led to a plethora of approaches based on reinforcement learning, inductive logic programming and graph embeddings. In particular, rule-based KBC has led to interpretable rules while being comparable in performance with graph embeddings. Even within rule-based KBC, there exist different approaches that lead to rules of varying quality and previous work has not always been precise in highlighting these differences. Another issue that plagues most rule-based KBC is the non-uniformity of relation paths: some relation sequences occur in very few paths while others appear very frequently. In this paper, we show that not all rule-based KBC models are the same and propose two distinct approaches that learn in one case: 1) a mixture of relations and the other 2) a mixture of paths. When implemented on top of neuro-symbolic AI, which learns rules by extending Boolean logic to real-valued logic, the latter model leads to superior KBC accuracy outperforming state-of-the-art rule-based KBC by 2-10% in terms of mean reciprocal rank. Furthermore, to address the non-uniformity of relation paths, we combine rule-based KBC with graph embeddings thus improving our results even further and achieving the best of both worlds.

Gaetano Rossiello, Nandana Mihindukulasooriya, Ibrahim Abdelaziz et al.

Relation linking is essential to enable question answering over knowledge bases. Although there are various efforts to improve relation linking performance, the current state-of-the-art methods do not achieve optimal results, therefore, negatively impacting the overall end-to-end question answering performance. In this work, we propose a novel approach for relation linking framing it as a generative problem facilitating the use of pre-trained sequence-to-sequence models. We extend such sequence-to-sequence models with the idea of infusing structured data from the target knowledge base, primarily to enable these models to handle the nuances of the knowledge base. Moreover, we train the model with the aim to generate a structured output consisting of a list of argument-relation pairs, enabling a knowledge validation step. We compared our method against the existing relation linking systems on four different datasets derived from DBpedia and Wikidata. Our method reports large improvements over the state-of-the-art while using a much simpler model that can be easily adapted to different knowledge bases.

Ibrahim Abdelaziz, Maxwell Crouse, Bassem Makni et al.

Traditional automated theorem provers have relied on manually tuned heuristics to guide how they perform proof search. Recently, however, there has been a surge of interest in the design of learning mechanisms that can be integrated into theorem provers to improve their performance automatically. In this work, we introduce TRAIL, a deep learning-based approach to theorem proving that characterizes core elements of saturation-based theorem proving within a neural framework. TRAIL leverages (a) an effective graph neural network for representing logical formulas, (b) a novel neural representation of the state of a saturation-based theorem prover in terms of processed clauses and available actions, and (c) a novel representation of the inference selection process as an attention-based action policy. We show through a systematic analysis that these components allow TRAIL to significantly outperform previous reinforcement learning-based theorem provers on two standard benchmark datasets (up to 36% more theorems proved). In addition, to the best of our knowledge, TRAIL is the first reinforcement learning-based approach to exceed the performance of a state-of-the-art traditional theorem prover on a standard theorem proving benchmark (solving up to 17% more problems).

Francois Luus, Prithviraj Sen, Pavan Kapanipathi et al.

Answering logical queries over incomplete knowledge bases is challenging because: 1) it calls for implicit link prediction, and 2) brute force answering of existential first-order logic queries is exponential in the number of existential variables. Recent work of query embeddings provides fast querying, but most approaches model set logic with closed regions, so lack negation. Query embeddings that do support negation use densities that suffer drawbacks: 1) only improvise logic, 2) use expensive distributions, and 3) poorly model answer uncertainty. In this paper, we propose Logic Embeddings, a new approach to embedding complex queries that uses Skolemisation to eliminate existential variables for efficient querying. It supports negation, but improves on density approaches: 1) integrates well-studied t-norm logic and directly evaluates satisfiability, 2) simplifies modeling with truth values, and 3) models uncertainty with truth bounds. Logic Embeddings are competitively fast and accurate in query answering over large, incomplete knowledge graphs, outperform on negation queries, and in particular, provide improved modeling of answer uncertainty as evidenced by a superior correlation between answer set size and embedding entropy.

Pavan Kapanipathi, Ibrahim Abdelaziz, Srinivas Ravishankar et al.

Knowledge base question answering (KBQA)is an important task in Natural Language Processing. Existing approaches face significant challenges including complex question understanding, necessity for reasoning, and lack of large end-to-end training datasets. In this work, we propose Neuro-Symbolic Question Answering (NSQA), a modular KBQA system, that leverages (1) Abstract Meaning Representation (AMR) parses for task-independent question understanding; (2) a simple yet effective graph transformation approach to convert AMR parses into candidate logical queries that are aligned to the KB; (3) a pipeline-based approach which integrates multiple, reusable modules that are trained specifically for their individual tasks (semantic parser, entity andrelationship linkers, and neuro-symbolic reasoner) and do not require end-to-end training data. NSQA achieves state-of-the-art performance on two prominent KBQA datasets based on DBpedia (QALD-9 and LC-QuAD1.0). Furthermore, our analysis emphasizes that AMR is a powerful tool for KBQA systems.

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.

Anshuman Mishra, Dhruvesh Patel, Aparna Vijayakumar et al.

In the recent past, Natural language Inference (NLI) has gained significant attention, particularly given its promise for downstream NLP tasks. However, its true impact is limited and has not been well studied. Therefore, in this paper, we explore the utility of NLI for one of the most prominent downstream tasks, viz. Question Answering (QA). We transform the one of the largest available MRC dataset (RACE) to an NLI form, and compare the performances of a state-of-the-art model (RoBERTa) on both these forms. We propose new characterizations of questions, and evaluate the performance of QA and NLI models on these categories. We highlight clear categories for which the model is able to perform better when the data is presented in a coherent entailment form, and a structured question-answer concatenation form, respectively.

Anshuman Mishra, Dhruvesh Patel, Aparna Vijayakumar et al.

In recent years, the Natural Language Inference (NLI) task has garnered significant attention, with new datasets and models achieving near human-level performance on it. However, the full promise of NLI -- particularly that it learns knowledge that should be generalizable to other downstream NLP tasks -- has not been realized. In this paper, we study this unfulfilled promise from the lens of two downstream tasks: question answering (QA), and text summarization. We conjecture that a key difference between the NLI datasets and these downstream tasks concerns the length of the premise; and that creating new long premise NLI datasets out of existing QA datasets is a promising avenue for training a truly generalizable NLI model. We validate our conjecture by showing competitive results on the task of QA and obtaining the best reported results on the task of Checking Factual Correctness of Summaries.

Zijun Cui, Pavan Kapanipathi, Kartik Talamadupula et al.

Knowledge graphs (KGs) are of great importance to many real world applications, but they generally suffer from incomplete information in the form of missing relations between entities. Knowledge graph completion (also known as relation prediction) is the task of inferring missing facts given existing ones. Most of the existing work is proposed by maximizing the likelihood of observed instance-level triples. Not much attention, however, is paid to the ontological information, such as type information of entities and relations. In this work, we propose a type-augmented relation prediction (TaRP) method, where we apply both the type information and instance-level information for relation prediction. In particular, type information and instance-level information are encoded as prior probabilities and likelihoods of relations respectively, and are combined by following Bayes' rule. Our proposed TaRP method achieves significantly better performance than state-of-the-art methods on four benchmark datasets: FB15K, FB15K-237, YAGO26K-906, and DB111K-174. In addition, we show that TaRP achieves significantly improved data efficiency. More importantly, the type information extracted from a specific dataset can generalize well to other datasets through the proposed TaRP model.

Nandana Mihindukulasooriya, Gaetano Rossiello, Pavan Kapanipathi et al.

Knowledgebase question answering systems are heavily dependent on relation extraction and linking modules. However, the task of extracting and linking relations from text to knowledgebases faces two primary challenges; the ambiguity of natural language and lack of training data. To overcome these challenges, we present SLING, a relation linking framework which leverages semantic parsing using Abstract Meaning Representation (AMR) and distant supervision. SLING integrates multiple relation linking approaches that capture complementary signals such as linguistic cues, rich semantic representation, and information from the knowledgebase. The experiments on relation linking using three KBQA datasets; QALD-7, QALD-9, and LC-QuAD 1.0 demonstrate that the proposed approach achieves state-of-the-art performance on all benchmarks.

Keerthiram Murugesan, Mattia Atzeni, Pushkar Shukla et al.

In this paper, we consider the recent trend of evaluating progress on reinforcement learning technology by using text-based environments and games as evaluation environments. This reliance on text brings advances in natural language processing into the ambit of these agents, with a recurring thread being the use of external knowledge to mimic and better human-level performance. We present one such instantiation of agents that use commonsense knowledge from ConceptNet to show promising performance on two text-based environments.

Kshitij Fadnis, Kartik Talamadupula, Pavan Kapanipathi et al.

In this paper, we introduce the problem of knowledge graph contextualization -- that is, given a specific NLP task, the problem of extracting meaningful and relevant sub-graphs from a given knowledge graph. The task in the case of this paper is the textual entailment problem, and the context is a relevant sub-graph for an instance of the textual entailment problem -- where given two sentences p and h, the entailment relationship between them has to be predicted automatically. We base our methodology on finding paths in a cost-customized external knowledge graph, and building the most relevant sub-graph that connects p and h. We show that our path selection mechanism to generate sub-graphs not only reduces noise, but also retrieves meaningful information from large knowledge graphs. Our evaluation shows that using information on entities as well as the relationships between them improves on the performance of purely text-based systems.

Maxwell Crouse, Ibrahim Abdelaziz, Bassem Makni et al.

Automated theorem provers have traditionally relied on manually tuned heuristics to guide how they perform proof search. Deep reinforcement learning has been proposed as a way to obviate the need for such heuristics, however, its deployment in automated theorem proving remains a challenge. In this paper we introduce TRAIL, a system that applies deep reinforcement learning to saturation-based theorem proving. TRAIL leverages (a) a novel neural representation of the state of a theorem prover and (b) a novel characterization of the inference selection process in terms of an attention-based action policy. We show through systematic analysis that these mechanisms allow TRAIL to significantly outperform previous reinforcement-learning-based theorem provers on two benchmark datasets for first-order logic automated theorem proving (proving around 15% more theorems).

Pavan Kapanipathi, Veronika Thost, Siva Sankalp Patel et al.

Textual entailment is a fundamental task in natural language processing. Most approaches for solving the problem use only the textual content present in training data. A few approaches have shown that information from external knowledge sources like knowledge graphs (KGs) can add value, in addition to the textual content, by providing background knowledge that may be critical for a task. However, the proposed models do not fully exploit the information in the usually large and noisy KGs, and it is not clear how it can be effectively encoded to be useful for entailment. We present an approach that complements text-based entailment models with information from KGs by (1) using Personalized PageR- ank to generate contextual subgraphs with reduced noise and (2) encoding these subgraphs using graph convolutional networks to capture KG structure. Our technique extends the capability of text models exploiting structural and semantic information found in KGs. We evaluate our approach on multiple textual entailment datasets and show that the use of external knowledge helps improve prediction accuracy. This is particularly evident in the challenging BreakingNLI dataset, where we see an absolute improvement of 5-20% over multiple text-based entailment models.

Maxwell Crouse, Achille Fokoue, Maria Chang et al.

Machine learning systems regularly deal with structured data in real-world applications. Unfortunately, such data has been difficult to faithfully represent in a way that most machine learning techniques would expect, i.e. as a real-valued vector of a fixed, pre-specified size. In this work, we introduce a novel approach that compiles structured data into a satisfiability problem which has in its set of solutions at least (and often only) the input data. The satisfiability problem is constructed from constraints which are generated automatically a priori from a given signature, thus trivially allowing for a bag-of-words-esque vector representation of the input to be constructed. The method is demonstrated in two areas, automated reasoning and natural language processing, where it is shown to produce vector representations of natural-language sentences and first-order logic clauses that can be precisely translated back to their original, structured input forms.

Ryan Musa, Xiaoyan Wang, Achille Fokoue et al.

Open-domain question answering (QA) is an important problem in AI and NLP that is emerging as a bellwether for progress on the generalizability of AI methods and techniques. Much of the progress in open-domain QA systems has been realized through advances in information retrieval methods and corpus construction. In this paper, we focus on the recently introduced ARC Challenge dataset, which contains 2,590 multiple choice questions authored for grade-school science exams. These questions are selected to be the most challenging for current QA systems, and current state of the art performance is only slightly better than random chance. We present a system that rewrites a given question into queries that are used to retrieve supporting text from a large corpus of science-related text. Our rewriter is able to incorporate background knowledge from ConceptNet and -- in tandem with a generic textual entailment system trained on SciTail that identifies support in the retrieved results -- outperforms several strong baselines on the end-to-end QA task despite only being trained to identify essential terms in the original source question. We use a generalizable decision methodology over the retrieved evidence and answer candidates to select the best answer. By combining query rewriting, background knowledge, and textual entailment our system is able to outperform several strong baselines on the ARC dataset.

Xiaoyan Wang, Pavan Kapanipathi, Ryan Musa et al.

Natural Language Inference (NLI) is fundamental to many Natural Language Processing (NLP) applications including semantic search and question answering. The NLI problem has gained significant attention thanks to the release of large scale, challenging datasets. Present approaches to the problem largely focus on learning-based methods that use only textual information in order to classify whether a given premise entails, contradicts, or is neutral with respect to a given hypothesis. Surprisingly, the use of methods based on structured knowledge -- a central topic in artificial intelligence -- has not received much attention vis-a-vis the NLI problem. While there are many open knowledge bases that contain various types of reasoning information, their use for NLI has not been well explored. To address this, we present a combination of techniques that harness knowledge graphs to improve performance on the NLI problem in the science questions domain. We present the results of applying our techniques on text, graph, and text-to-graph based models, and discuss implications for the use of external knowledge in solving the NLI problem. Our model achieves the new state-of-the-art performance on the NLI problem over the SciTail science questions dataset.

Michael Boratko, Harshit Padigela, Divyendra Mikkilineni et al.

The recent work of Clark et al. introduces the AI2 Reasoning Challenge (ARC) and the associated ARC dataset that partitions open domain, complex science questions into an Easy Set and a Challenge Set. That paper includes an analysis of 100 questions with respect to the types of knowledge and reasoning required to answer them; however, it does not include clear definitions of these types, nor does it offer information about the quality of the labels. We propose a comprehensive set of definitions of knowledge and reasoning types necessary for answering the questions in the ARC dataset. Using ten annotators and a sophisticated annotation interface, we analyze the distribution of labels across the Challenge Set and statistics related to them. Additionally, we demonstrate that although naive information retrieval methods return sentences that are irrelevant to answering the query, sufficient supporting text is often present in the (ARC) corpus. Evaluating with human-selected relevant sentences improves the performance of a neural machine comprehension model by 42 points.

Q Vera Liao, Biplav Srivastava, Pavan Kapanipathi

Dialog is a natural modality for interaction between customers and businesses in the service industry. As customers call up the service provider, their interactions may be routine or extraordinary. We believe that these interactions, when seen as dialogs, can be analyzed to obtain a better understanding of customer needs and how to efficiently address them. We introduce the idea of a dialog complexity measure to characterize multi-party interactions, propose a general data-driven method to calculate it, use it to discover insights in public and enterprise dialog datasets, and demonstrate its beneficial usage in facilitating better handling of customer requests and evaluating service agents.