Exploring Generative Process Reward Modeling for Semi-Structured Data: A Case Study of Table Question Answering
On the map: LLM reasoning → Reinforcement learning for reasoning
Abstract (arXiv)
Process reward models (PRMs) enhance complex reasoning in large language models (LLMs) by evaluating candidate solutions step-by-step and selecting answers based on aggregated step scores. While effective in domains such as mathematics, their applicability to tasks involving semi-structured data, like table question answering (TQA), remains unexplored. TQA poses unique challenges for PRMs, including abundant irrelevant information, loosely connected reasoning steps, and domain-specific reasoning. This work presents the first systematic study of PRMs for TQA. We evaluate state-of-the-art generative PRMs on TQA from both answer and step perspectives. Results show that PRMs that combine textual and code verification can aid solution selection but struggle to generalize to out-of-domain data. Analysis reveals a weak correlation between performance in step-level verification and answer accuracy, possibly stemming from weak step dependencies and loose causal links. Our findings highlight limitations of current PRMs on TQA and offer valuable insights for building more robust, process-aware verifiers.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning. Builds on 15 of its 27 references that are in the map. First posted on arXiv 23 Oct 2025.
Builds on
15 of 27 references.
Closest ancestors · 6
Same direction, newest first
- 2025GenPRM: Scaling Test-Time Compute of Process Reward Models via Generative Reasoning1 Apr
- 2025Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domains31 Mar
- 2025R-PRM: Reasoning-Driven Process Reward Modeling27 Mar
- 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning22 Jan
- 2025The Lessons of Developing Process Reward Models in Mathematical Reasoning13 Jan
- 2023Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations14 Dec
Other references · 9
- 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
- 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
- 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
- 2024Qwen2.5 Technical Reportcited by 4,713
- 2024A Survey on LLM-as-a-Judgecited by 1,384
- 2019TabFact: A Large-scale Dataset for Table-based Fact Verificationcited by 425
- 2024Chain-of-Table: Evolving Tables in the Reasoning Chain for Table Understandingcited by 185
- 2025VisualPRM: An Effective Process Reward Model for Multimodal Reasoningcited by 134
- 2024Efficient Multi-Agent Collaboration with Tool Use for Online Planning in Complex Table Question Answeringcited by 13
Built on by
Citing papers per year: 2026: 2.
2 papers.
Similar, not cited
Close in wording, no citation either way
- 2024Seek and Solve Reasoning for Table Question Answering
- 2025A Survey of Process Reward Models: From Outcome Signals to Process Supervisions for Large Language Models
- 2025From Mathematical Reasoning to Code: Generalization of Process Reward Models in Test-Time Scaling
- 2026Process Reward Models Meet Planning: Generating Precise and Scalable Datasets for Step-Level Rewards
- 2023Let's reward step by step: Step-Level reward model as the Navigators for Reasoning
- 2026Process Reward Agents for Steering Knowledge-Intensive Reasoning
- 2025Linking Process to Outcome: Conditional Reward Modeling for LLM Reasoning
- 2026Off-the-Shelf LLMs as Process Scorers: Training-Free Alternative to PRMs for Mathematical Reasoning