18 papers
Training Reasoning Models to Generate Diverse Scientific Ideas
Manya Wadhwa, Pranav Narayanan Venkit, Hiroaki Hayashi +3
One of the first steps of a research pipeline is generating ideas. Large Language Models (LLMs) are increasingly used to automate this step, but most of this work optimizes the qua…
Agentic Self-Awareness: Aligning Knowing, Acting, and Holding
Qinglin Chen, Hiroaki Hayashi, Jiaxin Zhang +1
Frontier AI agents increasingly act in environments where seemingly solvable tasks may lack prerequisites, contain conflicting requirements, or reveal blockers only during executio…
Prospective Hindsight: Self-Calibrating Reinforcement Learning via Prediction-Reality Gaps
Jiaxin Zhang, Xiangyu Peng, Qinglin Chen +3
Reinforcement learning for long-horizon agents relies on purely retrospective training signals: credit is assigned only after observing environmental consequences, leaving the agen…
Fork Where the Model Changes Its Mind: Belief-Shift Branching for Tree-Structured Reinforcement Learning
Bin Lei, Yu Li, Prafulla Kumar Choubey +7
Tree-structured rollouts give critic-free reinforcement learning with verifiable rewards (RLVR) step-level credit: fork a chain at an intermediate point, and sibling outcome differ…
On the Fragility of Self-Improving Agents: Variance, Task Order, and Underspecification
Qinyuan Ye, Yu Li, Yada Pruksachatkun +2
Memory-based self-improving agents--those that learn from an online stream of tasks and improve over time by maintaining a textual memory bank--have shown great promise in recent l…
Belief Cascades Drive Persuasion in LLM Agent Networks
Haoyi Qiu, Genglin Liu, Pranav Narayanan Venkit +4
Multi-agent LLM systems increasingly debate answers, coordinate research, simulate users, and mediate information flows, making agent-to-agent persuasion a basic but undermeasured…
Best Friends, Not Forever: Evaluating Long-Horizon Persona Collapse and Behavioral Drift in AI Companions
Pranav Narayanan Venkit, Akshara Prabhakar, Yu Li +2
As AI companions increasingly mediate repeated social interaction, users may rely on a stable role and shared history, yet locally acceptable replies do not ensure that either pers…
Speculate with Memory: Lossless Acceleration for LLM Agents
Yu Li, Qinyuan Ye, Prafulla Kumar Choubey +2
Speculative execution accelerates LLM agents by using a smaller, cheaper model to predict and pre-launch the next step while the environment is idle. However, existing speculators…
GTA: Generating Long-Horizon Tasks for Web Agents at Scale
Tenghao Huang, Kung-Hsiang Huang, Prafulla Kumar Choubey +4
Web agents, which couple language models with browsing and tool-use capabilities, show promise as open web assistants. Yet progress is increasingly limited by the lack of scalable,…
Dont Stop Early: Scalable Enterprise Deep Research with Controlled Information Flow and Evidence-Aware Termination
Prafulla Kumar Choubey, Kung-Hsiang Huang, Pranav Narayanan Venkit +5
Enterprise deep research often fails to produce decision-ready reports due to uneven information coverage, context explosion, and premature stopping. We propose a scalable Enterpri…
The Illusion of Certainty: Decoupling Capability and Calibration in On-Policy Distillation
Jiaxin Zhang, Xiangyu Peng, Qinglin Chen +3
On-policy distillation (OPD) is an increasingly important paradigm for post-training language models. However, we identify a pervasive Scaling Law of Miscalibration: while OPD effe…
MTA-Agent: An Open Recipe for Multimodal Deep Search Agents
Xiangyu Peng, Can Qin, An Yan +4
Multimodal large language models (MLLMs) have demonstrated strong capabilities in visual understanding, yet they remain limited in complex, multi-step reasoning that requires deep…
Agentic Confidence Calibration
Jiaxin Zhang, Caiming Xiong, Chien-Sheng Wu
AI agents are rapidly advancing from passive language models to autonomous systems executing complex, multi-step tasks. Yet their overconfidence in failure remains a fundamental ba…
Agentic Uncertainty Quantification
Jiaxin Zhang, Prafulla Kumar Choubey, Kung-Hsiang Huang +2
Although AI agents have demonstrated impressive capabilities in long-horizon reasoning, their reliability is severely hampered by the ``Spiral of Hallucination,'' where early epist…
From Passive Metric to Active Signal: The Evolving Role of Uncertainty Quantification in Large Language Models
Jiaxin Zhang, Wendi Cui, Zhuohang Li +4
While Large Language Models (LLMs) show remarkable capabilities, their unreliability remains a critical barrier to deployment in high-stakes domains. This survey charts a functiona…
Seeing is Believing? Evaluating Vision-Language Model Susceptibility in Agent-to-Agent Multimodal Persuasion
Haoyi Qiu, Yilun Zhou, Pranav Narayanan Venkit +4
As autonomous agents increasingly interact, they inevitably attempt to influence one another. While prior work in text-only settings has explored the dynamics of Agent-to-Agent (A2…
GUI-KV: Efficient GUI Agents via KV Cache with Spatio-Temporal Awareness
Kung-Hsiang Huang, Haoyi Qiu, Yutong Dai +2
Graphical user interface (GUI) agents built on vision-language models have emerged as a promising approach to automate human-computer workflows. However, they also face the ineffic…
Nudging the Boundaries of LLM Reasoning
Justin Chih-Yao Chen, Becky Xiangyu Peng, Prafulla Kumar Choubey +4
Current online reinforcement learning (RL) algorithms like GRPO share a key limitation in LLM reasoning: they cannot learn from problems that are "unsolvable" to the model. In othe…