18 papers
GameSmith: Scaling Zero-Sum Self-Play with Synthesised Games
Weixian Waylon Li, Jiaxin Zhang, Han Zhou +4
Post-training large language models (LLMs) on two-player zero-sum games provides a verifiable reward, and the gains have been shown to transfer to general reasoning. Existing pipel…
SAVLA: Symmetry-Aware Vision-Language-Action Models for Robotic Manipulation
Junle Li, Weixian Waylon Li, Fuxiang Wu +2
Vision-language-action (VLA) models have become the dominant paradigm for language-conditioned robot manipulation. However, although images and language instructions inherently enc…
Answer-Level Trust Selection for Physical Vision-Language Reasoning
Rongyu Yu, Ke Niu, Fengxiang He
Vision-language models (VLMs) can estimate physical quantities such as duration, speed, and acceleration from visual observations, but existing benchmarks primarily assess overall…
Distributed Team Orchestration via Supervisor Networks: Convergence, Optimality, and Resilience
Juntian Zhu, Guanpu Chen, Tongtian Zhu +3
In this paper, we study zero-sum potential team games with a supervisor network, where agents rely on supervisor-provided belief information rather than accurate common beliefs. Th…
LLM-OSDA: An Optimal-Stopping Dynamic Auction for Native Advertising in Multi-Turn LLM Conversations
Yan Fang, Jialin Chen, Chun Gan +5
LLM-native advertising embeds sponsored content directly into model-generated responses, shifting the unit of sale from a fixed slot to a moment within an evolving conversation. Ex…
DeXposure-Claw: An Agentic System for DeFi Risk Supervision
Aijie Shu, Bowei Chen, Wenbin Wu +2
Decentralized finance exposes supervisors to fast-moving, networked credit risks. General-purpose LLM agents fit this setting poorly: they over-read weak evidence and recommend hig…
Beyond Global Divergences: A Local-Mass Perspective on Bayesian Inference
Hanli Xu, Fengxiang He, Sarat Moka
Global objectives, such as KL divergence and ELBO, are widely used in Bayesian inference for measuring distributional discrepancy. This paper studies distributional ``local-mass be…
DeepSWIP: Quotient-WMC Counterfactuals for Neural Probabilistic Logic Programs
Saimun Habib, Vaishak Belle, Fengxiang He
Neurosymbolic systems such as DeepProbLog combine neural perception with probabilistic logic, but standard inference is associational. Counterfactual reasoning additionally require…
In LLM Reasoning, there is Irrationality on top of Value Misalignment
Kejiang Qian, Fengxiang He
Significant progress has been made in aligning LLMs with target value functions. We argue that, even when an LLM has been well aligned in (post-)training, it may still fail to maxi…
PEAR: Permutation-Equivariant Adaptive Routing Multi-Agent Debate
Yang Feng, Ziwei Xu, Xia Hu +1
Multi-agent debate improves the reliability of large language models (LLMs) through iterative peer critiques. However, fixed topologies often introduce persistent positional biases…
AlphaMemo: Structured Search-Process Memory for Self-Evolving Alpha Mining Agents
Hang Yu, Zifan Zheng, Jeff Z. Pan +4
LLM agents are promising for alpha mining via combining financial priors, symbolic reasoning, executable factor generation, and feedback-driven refinement. Yet, they face a combina…
Generalisation of RLHF under Reward Shift and Clipped KL Regularisation
Kenton Tang, Yuzhu Chen, Fengxiang He
Alignment and adaptation in large language models heavily rely on reinforcement learning from human feedback (RLHF); yet, theoretical understanding of its generalisability remains…
PRISM: Parallel Reward Integration with Symmetry for MORL
Finn van der Knaap, Kejiang Qian, Zheng Xu +1
This work studies heterogeneous Multi-Objective Reinforcement Learning (MORL), where objectives can differ sharply in temporal frequency. Such heterogeneity allows dense objectives…
Rationality Measurement and Theory for Reinforcement Learning Agents
Kejiang Qian, Amos Storkey, Fengxiang He
This paper proposes a suite of rationality measures and associated theory for reinforcement learning agents, a property increasingly critical yet rarely explored. We define an acti…
DeXposure-FM: A Time-series, Graph Foundation Model for Credit Exposures and Stability on Decentralized Financial Networks
Aijie Shu, Wenbin Wu, Gbenga Ibikunle +1
Credit exposure in Decentralized Finance (DeFi) is often implicit and token-mediated, creating a dense web of inter-protocol dependencies. Thus, a shock to one token may result in…
DeXposure: A Dataset and Benchmarks for Inter-protocol Credit Exposure in Decentralized Financial Networks
Wenbin Wu, Kejiang Qian, Alexis Lui +5
We curate the DeXposure dataset, the first large-scale dataset for inter-protocol credit exposure in decentralized financial networks, covering global markets of 43.7 million entri…
The Current State of AI Bias Bounties: An Overview of Existing Programmes and Research
Sergej Kucenko, Nathaniel Dennler, Fengxiang He
Current bias evaluation methods rarely engage with communities impacted by AI systems. Inspired by bug bounties, bias bounties have been proposed as a reward-based method that invo…
When a Reinforcement Learning Agent Encounters Unknown Unknowns
Juntian Zhu, Miguel de Carvalho, Zhouwang Yang +1
An AI agent might surprisingly find she has reached an unknown state which she has never been aware of -- an unknown unknown. We mathematically ground this scenario in reinforcemen…