Vishal Pallagani

Riju Marwah, Ritvik Garimella, Vishal Pallagani et al.

Autoregressive language models frequently degrade during long-horizon generation, producing repetitive text, losing instruction adherence, and exhibiting unstable entropy. Despite the prevalence of these failures, practitioners lack online diagnostics to detect them in real-time as they occur. We formalize this degradation as cognitive fatigue, a measurable generation-time state characterized by decay in attention to the original prompt, representational drift, and entropy miscalibration. We introduce the Fatigue Index (FI), a lightweight, model-agnostic diagnostic that aggregates these three signals under explicit axioms (monotonicity, boundedness, interpretability) enabling reliable runtime monitoring. Across nine models (1B-13B parameters), FI trajectories exhibit structured temporal dynamics, predict task degradation (AUROC = 0.95) and repetition (Spearman rho = 0.94), and reveal non-monotonic scaling behavior: instruction-tuned models below 3B exhibit faster collapse than base models, with this trend reversing at 7B. Stress analyses further show that FI onset accelerates under longer contexts, middle-positioned evidence, and reduced numerical precision. These results establish cognitive fatigue as a coherent and measurable phenomenon, and position FI as a principled tool for runtime reliability monitoring in production LLM systems.

Vishal Pallagani, Bharath Muppasani, Keerthiram Murugesan et al.

Large Language Models (LLMs) have been the subject of active research, significantly advancing the field of Natural Language Processing (NLP). From BERT to BLOOM, LLMs have surpassed state-of-the-art results in various natural language tasks such as question answering, summarization, and text generation. Many ongoing efforts focus on understanding LLMs' capabilities, including their knowledge of the world, syntax, and semantics. However, extending the textual prowess of LLMs to symbolic reasoning has been slow and predominantly focused on tackling problems related to the mathematical field. In this paper, we explore the use of LLMs for automated planning - a branch of AI concerned with the realization of action sequences (plans) to achieve a goal, typically executed by intelligent agents, autonomous robots, and unmanned vehicles. We introduce Plansformer; an LLM fine-tuned on planning problems and capable of generating plans with favorable behavior in terms of correctness and length with reduced knowledge-engineering efforts. We also demonstrate the adaptability of Plansformer in solving different planning domains with varying complexities, owing to the transfer learning abilities of LLMs. For one configuration of Plansformer, we achieve ~97% valid plans, out of which ~95% are optimal for Towers of Hanoi - a puzzle-solving domain.

Bharath Muppasani, Vishal Pallagani, Biplav Srivastava et al.

Ontologies are known for their ability to organize rich metadata, support the identification of novel insights via semantic queries, and promote reuse. In this paper, we consider the problem of automated planning, where the objective is to find a sequence of actions that will move an agent from an initial state of the world to a desired goal state. We hypothesize that given a large number of available planners and diverse planning domains; they carry essential information that can be leveraged to identify suitable planners and improve their performance for a domain. We use data on planning domains and planners from the International Planning Competition (IPC) to construct a planning ontology and demonstrate via experiments in two use cases that the ontology can lead to the selection of promising planners and improving their performance using macros - a form of action ordering constraints extracted from planning ontology. We also make the planning ontology and associated resources available to the community to promote further research.

Francesco Fabiano, Vishal Pallagani, Marianna Bergamaschi Ganapini et al.

The concept of Artificial Intelligence has gained a lot of attention over the last decade. In particular, AI-based tools have been employed in several scenarios and are, by now, pervading our everyday life. Nonetheless, most of these systems lack many capabilities that we would naturally consider to be included in a notion of "intelligence". In this work, we present an architecture that, inspired by the cognitive theory known as Thinking Fast and Slow by D. Kahneman, is tasked with solving planning problems in different settings, specifically: classical and multi-agent epistemic. The system proposed is an instance of a more general AI paradigm, referred to as SOFAI (for Slow and Fast AI). SOFAI exploits multiple solving approaches, with different capabilities that characterize them as either fast or slow, and a metacognitive module to regulate them. This combination of components, which roughly reflects the human reasoning process according to D. Kahneman, allowed us to enhance the reasoning process that, in this case, is concerned with planning in two different settings. The behavior of this system is then compared to state-of-the-art solvers, showing that the newly introduced system presents better results in terms of generality, solving a wider set of problems with an acceptable trade-off between solving times and solution accuracy.

Kausik Lakkaraju, Sai Krishna Revanth Vuruma, Vishal Pallagani et al.

Increasingly powerful Large Language Model (LLM) based chatbots, like ChatGPT and Bard, are becoming available to users that have the potential to revolutionize the quality of decision-making achieved by the public. In this context, we set out to investigate how such systems perform in the personal finance domain, where financial inclusion has been an overarching stated aim of banks for decades. We asked 13 questions representing banking products in personal finance: bank account, credit card, and certificate of deposits and their inter-product interactions, and decisions related to high-value purchases, payment of bank dues, and investment advice, and in different dialects and languages (English, African American Vernacular English, and Telugu). We find that although the outputs of the chatbots are fluent and plausible, there are still critical gaps in providing accurate and reliable financial information using LLM-based chatbots.

Marianna B. Ganapini, Francesco Fabiano, Lior Horesh et al.

Nudging is a behavioral strategy aimed at influencing people's thoughts and actions. Nudging techniques can be found in many situations in our daily lives, and these nudging techniques can targeted at human fast and unconscious thinking, e.g., by using images to generate fear or the more careful and effortful slow thinking, e.g., by releasing information that makes us reflect on our choices. In this paper, we propose and discuss a value-based AI-human collaborative framework where AI systems nudge humans by proposing decision recommendations. Three different nudging modalities, based on when recommendations are presented to the human, are intended to stimulate human fast thinking, slow thinking, or meta-cognition. Values that are relevant to a specific decision scenario are used to decide when and how to use each of these nudging modalities. Examples of values are decision quality, speed, human upskilling and learning, human agency, and privacy. Several values can be present at the same time, and their priorities can vary over time. The framework treats values as parameters to be instantiated in a specific decision environment.

Bharath Muppasani, Vishal Pallagani, Biplav Srivastava et al.

Rubik's Cube (RC) is a well-known and computationally challenging puzzle that has motivated AI researchers to explore efficient alternative representations and problem-solving methods. The ideal situation for planning here is that a problem be solved optimally and efficiently represented in a standard notation using a general-purpose solver and heuristics. The fastest solver today for RC is DeepCubeA with a custom representation, and another approach is with Scorpion planner with State-Action-Space+ (SAS+) representation. In this paper, we present the first RC representation in the popular PDDL language so that the domain becomes more accessible to PDDL planners, competitions, and knowledge engineering tools, and is more human-readable. We then bridge across existing approaches and compare performance. We find that in one comparable experiment, DeepCubeA (trained with 12 RC actions) solves all problems with varying complexities, albeit only 78.5% are optimal plans. For the same problem set, Scorpion with SAS+ representation and pattern database heuristics solves 61.50% problems optimally, while FastDownward with PDDL representation and FF heuristic solves 56.50% problems, out of which 79.64% of the plans generated were optimal. Our study provides valuable insights into the trade-offs between representational choice and plan optimality that can help researchers design future strategies for challenging domains combining general-purpose solving methods (planning, reinforcement learning), heuristics, and representations (standard or custom).

Bharath Muppasani, Vishal Pallagani, Kausik Lakkaraju et al.

Chatbots, or bots for short, are multi-modal collaborative assistants that can help people complete useful tasks. Usually, when chatbots are referenced in connection with elections, they often draw negative reactions due to the fear of mis-information and hacking. Instead, in this paper, we explore how chatbots may be used to promote voter participation in vulnerable segments of society like senior citizens and first-time voters. In particular, we build a system that amplifies official information while personalizing it to users' unique needs transparently. We discuss its design, build prototypes with frequently asked questions (FAQ) election information for two US states that are low on an ease-of-voting scale, and report on its initial evaluation in a focus group. Our approach can be a win-win for voters, election agencies trying to fulfill their mandate and democracy at large.

Amit Sheth, Vishal Pallagani, Kaushik Roy

Generative AI, especially via Large Language Models (LLMs), has transformed content creation across text, images, and music, showcasing capabilities in following instructions through prompting, largely facilitated by instruction tuning. Instruction tuning is a supervised fine-tuning method where LLMs are trained on datasets formatted with specific tasks and corresponding instructions. This method systematically enhances the model's ability to comprehend and execute the provided directives. Despite these advancements, LLMs still face challenges in consistently interpreting complex, multi-step instructions and generalizing them to novel tasks, which are essential for broader applicability in real-world scenarios. This article explores why neurosymbolic AI offers a better path to enhance the instructability of LLMs. We explore the use a symbolic task planner to decompose high-level instructions into structured tasks, a neural semantic parser to ground these tasks into executable actions, and a neuro-symbolic executor to implement these actions while dynamically maintaining an explicit representation of state. We also seek to show that neurosymbolic approach enhances the reliability and context-awareness of task execution, enabling LLMs to dynamically interpret and respond to a wider range of instructional contexts with greater precision and flexibility.

Vishal Pallagani, Biplav Srivastava, Nitin Gupta

Text summarization is a well-studied problem that deals with deriving insights from unstructured text consumed by humans, and it has found extensive business applications. However, many real-life tasks involve generating a series of actions to achieve specific goals, such as workflows, recipes, dialogs, and travel plans. We refer to them as planning-like (PL) tasks noting that the main commonality they share is control flow information. which may be partially specified. Their structure presents an opportunity to create more practical summaries to help users make quick decisions. We investigate this observation by introducing a novel plan summarization problem, presenting a dataset, and providing a baseline method for generating PL summaries. Using quantitative metrics and qualitative user studies to establish baselines, we evaluate the plan summaries from our method and large language models. We believe the novel problem and dataset can reinvigorate research in summarization, which some consider as a solved problem.

Nitin Gupta, Vishal Pallagani, John A. Aydin et al.

Generalized planning studies the construction of solution strategies that generalize across families of planning problems sharing a common domain model, formally defined by a transition function $γ: S \times A \rightarrow S$. Classical approaches achieve such generalization through symbolic abstractions and explicit reasoning over $γ$. In contrast, recent Transformer-based planners, such as PlanGPT and Plansformer, largely cast generalized planning as direct action-sequence prediction, bypassing explicit transition modeling. While effective on in-distribution instances, these approaches typically require large datasets and model sizes, and often suffer from state drift in long-horizon settings due to the absence of explicit world-state evolution. In this work, we formulate generalized planning as a transition-model learning problem, in which a neural model explicitly approximates the successor-state function $\hatγ \approx γ$ and generates plans by rolling out symbolic state trajectories. Instead of predicting actions directly, the model autoregressively predicts intermediate world states, thereby learning the domain dynamics as an implicit world model. To study size-invariant generalization and sample efficiency, we systematically evaluate multiple state representations and neural architectures, including relational graph encodings. Our results show that learning explicit transition models yields higher out-of-distribution satisficing-plan success than direct action-sequence prediction in multiple domains, while achieving these gains with significantly fewer training instances and smaller models. This is an extended version of a short paper accepted at ICAPS 2026 under the same title.

Shweta Yadav, Vishal Pallagani, Amit Sheth

One of the cardinal tasks in achieving robust medical question answering systems is textual entailment. The existing approaches make use of an ensemble of pre-trained language models or data augmentation, often to clock higher numbers on the validation metrics. However, two major shortcomings impede higher success in identifying entailment: (1) understanding the focus/intent of the question and (2) ability to utilize the real-world background knowledge to capture the context beyond the sentence. In this paper, we present a novel Medical Knowledge-Enriched Textual Entailment framework that allows the model to acquire a semantic and global representation of the input medical text with the help of a relevant domain-specific knowledge graph. We evaluate our framework on the benchmark MEDIQA-RQE dataset and manifest that the use of knowledge enriched dual-encoding mechanism help in achieving an absolute improvement of 8.27% over SOTA language models. We have made the source code available here.

Vishal Pallagani, Kaushik Roy, Bharath Muppasani et al.

Automated Planning and Scheduling is among the growing areas in Artificial Intelligence (AI) where mention of LLMs has gained popularity. Based on a comprehensive review of 126 papers, this paper investigates eight categories based on the unique applications of LLMs in addressing various aspects of planning problems: language translation, plan generation, model construction, multi-agent planning, interactive planning, heuristics optimization, tool integration, and brain-inspired planning. For each category, we articulate the issues considered and existing gaps. A critical insight resulting from our review is that the true potential of LLMs unfolds when they are integrated with traditional symbolic planners, pointing towards a promising neuro-symbolic approach. This approach effectively combines the generative aspects of LLMs with the precision of classical planning methods. By synthesizing insights from existing literature, we underline the potential of this integration to address complex planning challenges. Our goal is to encourage the ICAPS community to recognize the complementary strengths of LLMs and symbolic planners, advocating for a direction in automated planning that leverages these synergistic capabilities to develop more advanced and intelligent planning systems.

Biplav Srivastava, Vishal Pallagani

Foundation Models (FMs) have revolutionized many areas of computing, including Automated Planning and Scheduling (APS). For example, a recent study found them useful for planning problems: plan generation, language translation, model construction, multi-agent planning, interactive planning, heuristics optimization, tool integration, and brain-inspired planning. Besides APS, there are many seemingly related tasks involving the generation of a series of actions with varying guarantees of their executability to achieve intended goals, which we collectively call planning-like (PL) tasks like business processes, programs, workflows, and guidelines, where researchers have considered using FMs. However, previous works have primarily focused on pre-trained, off-the-shelf FMs and optionally fine-tuned them. This paper discusses the need for a comprehensive FM for PL tasks from scratch and explores its design considerations. We argue that such an FM will open new and efficient avenues for PL problem-solving, just like LLMs are creating for APS.

Vedant Khandelwal, Vishal Pallagani, Biplav Srivastava et al.

Constraint Satisfaction Problems (CSPs) present significant challenges to artificial intelligence due to their intricate constraints and the necessity for precise solutions. Existing symbolic solvers are often slow, and prior research has shown that Large Language Models (LLMs) alone struggle with CSPs because of their complexity. To bridge this gap, we build upon the existing SOFAI architecture (SOFAI_v1), which adapts Daniel Kahneman's ''Thinking, Fast and Slow'' cognitive model to AI. Our enhanced architecture, SOFAI_v2, integrates refined metacognitive governance mechanisms to improve adaptability across complex domains, specifically tailored here for solving the graph coloring problem, a specific type of CSP. SOFAI_v2 combines a fast System 1 (S1), leveraging LLMs, with a deliberative System 2 (S2), governed by a metacognition module. S1's initial solutions, often limited by constraint adherence issues, are improved through targeted feedback and examples from metacognition, aligning S1 more closely with CSP requirements. If S1 fails to resolve the problem, metacognition strategically invokes S2, ensuring accurate and reliable solutions. Our empirical results demonstrate that SOFAI_v2 achieves a 10.5% higher success rate and is up to 30% faster than a traditional symbolic solver in solving graph coloring problems.

Vishal Pallagani, Nitin Gupta, John Aydin et al.

Understanding how data moves, transforms, and persists, known as data flow, is fundamental to reasoning in procedural tasks. Despite their fluency in natural and programming languages, large language models (LLMs), although increasingly being applied to decisions with procedural tasks, have not been systematically evaluated for their ability to perform data-flow reasoning. We introduce FABLE, an extensible benchmark designed to assess LLMs' understanding of data flow using structured, procedural text. FABLE adapts eight classical data-flow analyses from software engineering: reaching definitions, very busy expressions, available expressions, live variable analysis, interval analysis, type-state analysis, taint analysis, and concurrency analysis. These analyses are instantiated across three real-world domains: cooking recipes, travel routes, and automated plans. The benchmark includes 2,400 question-answer pairs, with 100 examples for each domain-analysis combination. We evaluate three types of LLMs: a reasoning-focused model (DeepSeek-R1 8B), a general-purpose model (LLaMA 3.1 8B), and a code-specific model (Granite Code 8B). Each model is tested using majority voting over five sampled completions per prompt. Results show that the reasoning model achieves higher accuracy, but at the cost of over 20 times slower inference compared to the other models. In contrast, the general-purpose and code-specific models perform close to random chance. FABLE provides the first diagnostic benchmark to systematically evaluate data-flow reasoning and offers insights for developing models with stronger procedural understanding.

Vishal Pallagani, Bharath Muppasani, Keerthiram Murugesan et al.

Automated planning is concerned with developing efficient algorithms to generate plans or sequences of actions to achieve a specific goal in a given environment. Emerging Large Language Models (LLMs) can answer questions, write high-quality programming code, and predict protein folding, showcasing their versatility in solving various tasks beyond language-based problems. In this paper, we aim to explore how LLMs can also be used for automated planning. To do so, we seek to answer four key questions. Firstly, we want to understand the extent to which LLMs can be used for plan generation. Secondly, we aim to identify which pre-training data is most effective in facilitating plan generation. Thirdly, we investigate whether fine-tuning or prompting is a more effective approach for plan generation. Finally, we explore whether LLMs are capable of plan generalization. By answering these questions, the study seeks to shed light on the capabilities of LLMs in solving complex planning problems and provide insights into the most effective approaches for using LLMs in this context.

Vishal Pallagani, Priyadharsini Ramamurthy, Vedant Khandelwal et al.

Food is not only a basic human necessity but also a key factor driving a society's health and economic well-being. As a result, the cooking domain is a popular use-case to demonstrate decision-support (AI) capabilities in service of benefits like precision health with tools ranging from information retrieval interfaces to task-oriented chatbots. An AI here should understand concepts in the food domain (e.g., recipes, ingredients), be tolerant to failures encountered while cooking (e.g., browning of butter), handle allergy-based substitutions, and work with multiple data modalities (e.g. text and images). However, the recipes today are handled as textual documents which makes it difficult for machines to read, reason and handle ambiguity. This demands a need for better representation of the recipes, overcoming the ambiguity and sparseness that exists in the current textual documents. In this paper, we discuss the construction of a machine-understandable rich recipe representation (R3), in the form of plans, from the recipes available in natural language. R3 is infused with additional knowledge such as information about allergens and images of ingredients, possible failures and tips for each atomic cooking step. To show the benefits of R3, we also present TREAT, a tool for recipe retrieval which uses R3 to perform multi-modal reasoning on the recipe's content (plan objects - ingredients and cooking tools), food preparation process (plan actions and time), and media type (image, text). R3 leads to improved retrieval efficiency and new capabilities that were hither-to not possible in textual representation.