1 citations · 1 across the 16 of their papers we have counts for
21 papers · 1 filter
CascadeLedger: Counterfactual Auditing of Information and Resource Integrity in LLM Cascades
Miaobo Hu, Shuhao Hu, Xiaobo Guo +4
We introduce CascadeLedger, a counterfactual execution-audit framework for budgeted language-model cascades. The protocol enforces call-before-read semantics, binds successful read…
FedRIC: Heterogeneity-Aware Federated Learning with Hard Deadline Constraints in Open RAN
Adhwaa Al-Chaab, Ayman Younis, Dario Pompili
Open Radio Access Networks (O-RANs) bring intelligence to disaggregated RANs through virtualized platforms and open interfaces, but the stringent latency budgets of the RAN Intelli…
Quantizer Redundancy and Exact Index Recoding in a Frozen Neural Audio Codec
Youssef Hemimy
Residual scalar quantization can produce highly dependent indices when successive stages reuse identical, unscaled grids. This case study examines that behavior in a frozen neural…
FSPO: Policy-Consistent Risk and Pareto-Feasible Control for Budgeted LLM RL Post-Training
Miaobo Hu, Shuhao Hu, Xiaobo Guo +4
Adaptive LLM reinforcement-learning post-training changes multiple training actuators online, including rollout temperature, group size, clipping, KL regularization, verifier alloc…
ROUTEAUDIT: Interaction-Aware Identification for Budgeted Multi-Verifier Routing
Miaobo Hu, Shuhao Hu, Xiaobo Guo +5
Adaptive multi-verifier systems are commonly compared through endpoint quality-cost gaps, even when the verifier catalog, availability, accounting, information filtration, or score…
FAER: Auditable Utility-Aligned Trajectory Replay for Language Model Post-Training
Miaobo Hu, Shuhao Hu, Xiaobo Guo +5
Replay selectors often rank cached trajectories by format feedback, confidence, freshness, or response length, although cache-level correctness and downstream learner utility are d…
ContractRL: Shielded Group-Relative Policy Optimization for Auditable Tool-Call Repair
Miaobo Hu, Shuhao Hu, Xiaobo Guo +5
Structured tool calls often fail after only a small number of fields violate a schema or an execution contract. Regenerating the complete object enlarges the action surface and mak…
Actions with Receipts: Jointly Binding Claims, Evidence, and Execution for Replayable Tool-Agent Auditing
Miaobo Hu, Shuhao Hu, Xiaobo Guo +5
Tool-using agents can expose citations and execution logs while leaving a critical association unaudited: whether the claim shown to a user is the claim emitted by the committed ex…
VStress: Correlation-Aware Auditing and Adaptive Budget Allocation for Repeated Verifiers
Miaobo Hu, Shuhao Hu, Xiaobo Guo +5
Repeated verifier calls are useful only when they contribute conditional information. We introduce VStress, an auditable replay contract, and VStress-CA, a correlation-aware alloca…
RSD-Poker: Structure-Adaptive and Shift-Robust Risk-Utility Certification for Residual Policies in Imperfect-Information Games
Miaobo Hu, Shuhao Hu, Xiaobo Guo +5
Residual policy adaptation provides a lightweight way to modify a strong reference policy, but a shared scale and a fixed subgroup partition can hide heterogeneous degradation and…
Audit-First VAPO: Risk-Certified Selective Updates under Imperfect Verification
Miaobo Hu, Shuhao Hu, Xiaobo Guo +5
Imperfect verifiers can assign a harmful update direction even when clipping and regularization bound its magnitude. We introduce Audit-First VAPO, which separates discrete directi…
OSCC: Certified Observation-Safe Coupling Optimization for Gradient-Noise Control in Imperfect-Information Learning
Miaobo Hu, Shuhao Hu, Xiaobo Guo +5
Coupled rollouts can reduce the noise of counterfactual action comparisons, but two issues prevent standard common-random-number constructions from serving as a general learning pr…
QUADS: Stabilizing NVFP4 Reinforcement Learning for MoE via QUantization-error Alignment across Dual Sides
Zhengyang Zhuge, Hao Yu, Xin Wang +4
Rollout generation is a major bottleneck in Reinforcement Learning (RL) for Mixture-of-Experts (MoE) Large Language Models, motivating low-precision rollout acceleration such as FP…
Survive or Collapse: The Asymmetric Roles of Data Gating and Reward Grounding in Self-Play RL
Sophia Xiao Pu, Zhaotian Weng, Chengzhi Liu +4
Self-play reinforcement learning trains language models on their own generated tasks, co-evolving a proposer and solver without human labels. Recent systems report strong reasoning…
ECPO: Evidence-Coupled Policy Optimization for Verifier-Audited Candidate Ranking
Miaobo Hu, Shuhao Hu, BoKun Wang +5
Ranking systems used for decision support should expose evidence that can be audited independently of the model's ranking scores. We study a closed-roster setting in which a policy…
Towards Context-Invariant Safety Alignment for Large Language Models
Yixu Wang, Yang Yao, Xin Wang +4
Preference-based post-training aligns LLMs with human intent, yet safety behavior often remains brittle. A model may refuse a harmful request in a standard prompt but comply when t…
AGPO: Adaptive Group Policy Optimization with Dual Statistical Feedback
Miaobo Hu, Shuhao Hu, Bokun Wang +5
Reinforcement learning improves LLM reasoning, but PPO/GRPO typically use fixed clipping and decoding temperature, which makes training brittle and tuning-heavy. We propose Adaptiv…
SAVER: Selective As-Needed Vision Evidence for Multimodal Information Extraction
Miaobo Hu, Shuhao Hu, Bokun Wang +5
Multimodal IE in social media is difficult because a post may attach multiple images that are weakly related, redundant, or even misleading with respect to the text. In this settin…
TAME: Test-Time Adversarial Prompt Tuning via Mixture-of-Experts for Vision-Language Models
Xin Wang, Yixu Wang, Jiaming Zhang +6
Large-scale pre-trained Vision-Language models (VLMs), such as CLIP, exhibit strong zero-shot generalization, yet remain highly vulnerable to imperceptible adversarial perturbation…
DarkLLM: Learning Language-Driven Adversarial Attacks with Large Language Models
Ye Sun, Xin Wang, Jiaming Zhang +7
While vision and multimodal foundation models underpin critical tasks from perception to complex reasoning, they remain highly vulnerable to adversarial attacks. However, tradition…
QuantVLA: Scale-Calibrated Post-Training Quantization for Vision-Language-Action Models
Jingxuan Zhang, Yunta Hsieh, Zhongwei Wan +5
Vision-language-action (VLA) models unify perception, language, and control for embodied agents but face significant challenges in practical deployment due to rapidly increasing co…