76 citations · 390 across the 60 of their papers we have counts for
62 papers
Correct Answers, Invalid Traces: What Verifiable Grade-School Math Reveals About Chain-of-Thought Traces
Ratish Puduppully, Pranabendu Misra, Paarth Iyer +3
Chain-of-thought traces are widely read as records of how models reach their answers, informing debugging, agent auditing, and claims about reasoning. Testing this interpretation i…
Masked Self-Distillation: Internalizing the Chain-of-Thought in Language Models
Durgesh Kalwar, Vardhan Palod, Subbarao Kambhampati +1
Large Reasoning Models produce long, explicit chains of intermediate steps before generating a final answer at inference time. These intermediate traces dominate latency, memory us…
Evaluating the False Trust Engendered by LLM Explanations
Vardhan Palod, Upasana Biswas, Subbarao Kambhampati
Large Language Models (LLMs) and Large Reasoning Models (LRMs) are increasingly used for critical tasks, yet they provide no guarantees about the correctness of their solutions. Us…
NestRL: A Nested Training Regime for Mutual Adaptation in Human-AI Teaming
Upasana Biswas, Durgesh Kalwar, Subbarao Kambhampati +1
Mutual adaptation is a central challenge in human-AI teaming, as humans naturally adjust their strategies in response to an AI agent's behavior. Existing approaches attempt to appr…
interwhen: A Generalizable Framework for Steering Reasoning Models with Test-time Verification
Vishak K Bhat, Prateek Chanda, Vijval Ekbote +6
Reasoning models produce long traces of intermediate decisions and tool calls, making test-time verification important for ensuring correctness. Existing approaches either verify o…
Magentic Marketplace: An Open-Source Environment for Studying Agentic Markets
Gagan Bansal, Wenyue Hua, Zezhou Huang +21
As LLM agents advance, they are increasingly mediating economic decisions, ranging from product discovery to transactions, on behalf of users. Such applications promise benefits bu…
Local Coherence or Global Validity? Investigating RLVR Traces in Math Domains
Soumya Rani Samineni, Durgesh Kalwar, Vardaan Gangal +2
Reinforcement Learning with Verifiable Rewards (RLVR)-based post-training of Large Language Models (LLMs) has been shown to improve accuracy on reasoning tasks and continues to att…
Performative Thinking? The Brittle Correlation Between CoT Length and Problem Complexity
Vardhan Palod, Karthik Valmeekam, Kaya Stechly +1
Intermediate token generation (ITG), where a model produces output before the solution, has been proposed as a method to improve the performance of language models on reasoning tas…
Do Cognitively Interpretable Reasoning Traces Improve LLM Performance?
Siddhant Bhambri, Upasana Biswas, Subbarao Kambhampati
Recent progress in reasoning-oriented Large Language Models (LLMs) has been driven by introducing Chain-of-Thought (CoT) traces, where models generate intermediate reasoning traces…
Interpretable Traces, Unexpected Outcomes: Investigating the Disconnect in Trace-Based Knowledge Distillation
Siddhant Bhambri, Upasana Biswas, Subbarao Kambhampati
Recent advances in reasoning-focused Large Language Models (LLMs) have introduced Chain-of-Thought (CoT) traces - intermediate reasoning steps generated before a final answer. Thes…
Beyond Semantics: The Unreasonable Effectiveness of Reasonless Intermediate Tokens
Karthik Valmeekam, Vardhan Palod, Kaya Stechly +2
Recent impressive results from large reasoning models have been interpreted as a triumph of Chain of Thought (CoT), and especially of the process of training on CoTs sampled from b…
RL in Name Only? Analyzing the Structural Assumptions in RL post-training for LLMs
Soumya Rani Samineni, Durgesh Kalwar, Karthik Valmeekam +2
Reinforcement learning based post-training of large language models (LLMs) has recently gained attention, particularly following the release of DeepSeek R1, which applied GRPO for…
Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces!
Subbarao Kambhampati, Karthik Valmeekam, Siddhant Bhambri +6
Intermediate token generation (ITG), where a model produces output before the solution, has become a standard method to improve the performance of language models on reasoning task…
Who is Helping Whom? Analyzing Inter-dependencies to Evaluate Cooperation in Human-AI Teaming
Upasana Biswas, Vardhan Palod, Siddhant Bhambri +1
State-of-the-art methods for Human-AI Teaming and Zero-shot Cooperation focus on task completion, i.e., task rewards, as the sole evaluation metric while being agnostic to how the…
Robust Planning with Compound LLM Architectures: An LLM-Modulo Approach
Atharva Gundawar, Karthik Valmeekam, Mudit Verma +1
Previous work has attempted to boost Large Language Model (LLM) performance on planning and scheduling tasks through a variety of prompt engineering techniques. While these methods…
Planning in Strawberry Fields: Evaluating and Improving the Planning and Scheduling Capabilities of LRM o1
Karthik Valmeekam, Kaya Stechly, Atharva Gundawar +1
The ability to plan a course of action that achieves a desired state of affairs has long been considered a core competence of intelligent agents and has been an integral part of AI…
LLMs Still Can't Plan; Can LRMs? A Preliminary Evaluation of OpenAI's o1 on PlanBench
Karthik Valmeekam, Kaya Stechly, Subbarao Kambhampati
The ability to plan a course of action that achieves a desired state of affairs has long been considered a core competence of intelligent agents and has been an integral part of AI…
Algorithmic Language Models with Neurally Compiled Libraries
Lucas Saldyt, Subbarao Kambhampati
Important tasks such as reasoning and planning are fundamentally algorithmic, meaning that solving them robustly requires acquiring true reasoning or planning algorithms, rather th…
Robust Planning with LLM-Modulo Framework: Case Study in Travel Planning
Atharva Gundawar, Mudit Verma, Lin Guan +3
As the applicability of Large Language Models (LLMs) extends beyond traditional text processing tasks, there is a burgeoning interest in their potential to excel in planning and re…
On the Brittle Foundations of ReAct Prompting for Agentic Large Language Models
Mudit Verma, Siddhant Bhambri, Subbarao Kambhampati
The reasoning abilities of Large Language Models (LLMs) remain a topic of debate. Some methods such as ReAct-based prompting, have gained popularity for claiming to enhance sequent…
Explainable Human-AI Interaction: A Planning Perspective
Sarath Sreedharan, Anagha Kulkarni, Subbarao Kambhampati
From its inception, AI has had a rather ambivalent relationship with humans -- swinging between their augmentation and replacement. Now, as AI technologies enter our everyday lives…
Extracting Heuristics from Large Language Models for Reward Shaping in Reinforcement Learning
Siddhant Bhambri, Amrita Bhattacharjee, Durgesh Kalwar +3
Reinforcement Learning (RL) suffers from sample inefficiency in sparse reward domains, and the problem is further pronounced in case of stochastic transitions. To improve the sampl…
Chain of Thoughtlessness? An Analysis of CoT in Planning
Kaya Stechly, Karthik Valmeekam, Subbarao Kambhampati
Large language model (LLM) performance on reasoning problems typically does not generalize out of distribution. Previous work has claimed that this can be mitigated with chain of t…
Can Large Language Models Reason and Plan?
Subbarao Kambhampati
While humans sometimes do show the capability of correcting their own erroneous guesses with self-critiquing, there seems to be no basis for that assumption in the case of LLMs.