arXiv subjects

Mingze Gao

Publications and source records attributed to Mingze Gao.

At least 19 recordsLinked to original sources

arXiv2026

Beyond Semantic IDs: Atomic Creator Tokens for Large-Scale Generative Recommendation

Recommending creators that a user does not yet follow is a cold-start problem: from hundreds of thousands of popular creators and only a short trail of prior engagements, the system must predict the next creator worth surfacing. The prevailing generative recommendation recipe tokenizes each creator into a \emph{Semantic ID} (SID) by quantizing a frozen embedding and trains a language model to generate SIDs. We show that this SID layer, not the language model, is the binding constraint. The creator-to-SID map is many-to-one, so colliding creators must be resolved by a popularity tie-break that caps creator-level recall and favors already-popular creators, and the frozen quantizer stops the training gradient from reaching the creator representation. We propose Atomic Creator Token (ACT), which gives every creator its own vocabulary token trained end-to-end. The map is bijective, so collisions are impossible and inference is a direct top-$k$ over creator logits. In a controlled comparison on production data in which the token layer is the only variable, ACT improves creator-level Recall@15 by 19.8 points ($1.84\times$) over a matched SID baseline, still leads when the SID codebook is enlarged $64\times$, and serves at ${\sim}2.7\times$ lower per-user latency. The advantage carries over to the public KuaiRand dataset, where ACT is the only method that beats a global-popularity baseline, provided each creator token sees enough training examples. In a live A/B test, ACT is directionally positive on every reported metric against both the SID-enhanced and the classic production rankers.

cs.AI↗

arXiv2026

Baseline and Incremental: User, Item, and Request Debiasing in Instagram Feed Ranking

A feed ranker scores candidates with a value model, a weighted sum of engagement predictions. Each prediction conflates a baseline, what this viewer, item or request would produce anyway, with the incremental amount by which this candidate exceeds it. The literature mostly removes the baseline. We argue that a multi-signal value model should keep the raw prediction and add the incremental part as a separately weighted term. On Instagram Feed, three debiasing families deployed independently, for viewer habit, item popularity and request context, are one parameterised operator, and a design procedure picks a point in it. Media-type (format) bias goes instead to a debiased label chosen by the same procedure. In a controlled four-arm ablation of deployed Feed terms (neither term, each alone, both), the two-term arm beats both single-term arms at production weights on point estimates in 11 of 13 judged (term, metric) pairs. Of the 26 direct contrasts, 17 are significantly for it and only 1 against it, on impressions. The ablation also suggests that the offline diagnostic should screen signals, not rank them. Such terms have served Instagram Feed production for most of a year. Beyond Feed, Meta, whose apps serve billions of people, has adopted the operator across surfaces, apps and both ranking stages. Its launches there drive sessions and daily active users, the core visitation and retention metrics.

cs.IR↗

arXiv2026

An LLM-powered Agentic Recommendation System for Connected TV Content Discovery

Recommendation systems, from traditional multi-stage to recent unified generative architectures, face challenges in incorporating diverse contextual signals, such as trending topics, breaking news, cultural events, and cross-surface user activities, into their ranking pipelines. These systems are designed to consume structured behavioral signals with consistent schemas, and lack the reasoning capability to naturally process unstructured or heterogeneously formatted contextual information. Incorporating such signals typically requires feature engineering, bespoke data pipelines, and carefully tuned heuristics. In this paper, we present an LLM-powered agentic recommendation system designed for Connected TV (CTV) content discovery that addresses these limitations. Our system leverages the reasoning capabilities of large language models to naturally process and synthesize diverse signals across varying schemas and structures, eliminating much of the manual integration inherent in traditional ranking and retrieval systems. Recognizing that current LLM-based solutions still fall short of traditional machine learning models in several recommendation tasks, including retrieval efficiency, personalization precision, and scalability, we adopt an agentic architecture that orchestrates specialized components, allowing each sub-task to be handled by the most suitable method, whether LLM-based or traditional ML. The main contribution of this work is our engineering approach to successfully overcoming the practical limitations of enabling LLM for recommendation, particularly inference latency. We share insights from our work and discuss the trade-offs and lessons learned in building a hybrid system that combines the flexibility of LLMs with the performance of established recommendation techniques.

cs.IR↗

arXiv2026

PalmSpace: Towards a Versatile On-Palm Interaction Space through Unified Touch Modeling

As smart glasses and lightweight MR devices become increasingly practical, input remains a key challenge. The bare palm is an always-available, tactile, and proprioceptively accessible surface, but it has neither an explicit coordinate system nor embedded touch sensing. Prior on-palm systems typically expose isolated touch events, discrete regions, continuous trajectories, or task-specific gestures, limiting the palm's ability to support precise selection and gesture manipulation through a common input representation. We present PalmSpace, a wrist-worn infrared system that exposes mode-aware, body-referenced absolute input on the bare palm without per-user sensing calibration. At the interaction level, PalmSpace jointly represents contact occurrence, interaction mode, and palm-referenced absolute location; at the model level, it learns these coupled outputs through a shared real-time representation. In leave-one-participant-out evaluation with 17 participants, PalmSpace achieved 6.7 mm mean localization error, 98.9% contact detection accuracy, and 96.7% F1 for four-class interaction-state recognition. User studies further demonstrated absolute pointing and dragging, eyes-free digit input, and representative multi-finger controls including scrolling and pinch-based map manipulation. These results show that a morphologically variable bare palm can function as a transferable, mode-aware interaction surface.

cs.HC↗

arXiv2026

Decoding ML Decision: An Agentic Reasoning Framework for Large-Scale Ranking System

Modern large-scale ranking systems operate within a sophisticated landscape of competing objectives, operational constraints, and evolving product requirements. Progress in this domain is increasingly bottlenecked by the engineering context constraint: the arduous process of translating ambiguous product intent into reasonable, executable, verifiable hypotheses, rather than by modeling techniques alone. We present GEARS (Generative Engine for Agentic Ranking Systems), a framework that reframes ranking optimization as an autonomous discovery process within a programmable experimentation environment. Rather than treating optimization as static model selection, GEARS leverages Specialized Agent Skills to encapsulate ranking expert knowledge into reusable reasoning capabilities, enabling operators to steer systems via high-level intent vibe personalization. Furthermore, to ensure production reliability, the framework incorporates validation hooks to enforce statistical robustness and filter out brittle policies that overfit short-term signals. Experimental validation across diverse product surfaces demonstrates that GEARS consistently identifies superior, near-Pareto-efficient policies by synergizing algorithmic signals with deep ranking context while maintaining rigorous deployment stability.

cs.AI↗

arXiv2026

Persistent Negatives for Adversarial Black-Box On-Policy Distillation

Black-box On-Policy Distillation (OPD) seeks to improve a student from its own generations when the teacher provides sampled responses but not token probabilities. Adversarial distillation offers one route: it learns a discriminator over prompt-matched teacher and student responses and uses its score as the policy reward. However, sampling discriminator negatives from the latest student at each step couples the learned reward to a negative distribution that changes after every policy update. We address this moving-target problem with persistent-negative adversarial distillation, a live-pool method that replaces a fraction of each discriminator batch with historical, prompt-matched teacher--student comparisons. Under matched discriminator compute, historical comparisons train the discriminator, while GRPO remains on-policy with fresh student responses. Our analysis identifies the Bayes-optimal reward as a teacher-to-negative log-density ratio and, under explicit assumptions, shows how persistent negatives anchor the discriminator and reduce reward-estimation MSE relative to fresh-negative training. Across two student families, three judges, and four judged-chat benchmarks, persistent-negative adversarial distillation consistently improves performance over current methods at matched discriminator compute. It also yields smoother fresh-policy discriminator trajectories, with fewer below-chance dips. These findings identify the discriminator's negative distribution as an important design axis in black-box on-policy distillation.

cs.CL↗

arXiv2026

Earinter: A Closed-Loop System for Eating Pace Regulation with Just-in-Time Intervention Using Commodity Earbuds

Rapid eating is common yet difficult to regulate in situ, partly because people seldom notice pace changes and sustained self-monitoring is effortful. We present Earinter, a commodity-earbud-based closed-loop system that integrates in-the-wild sensing, real-time reasoning, and theory-grounded just-in-time (JIT) intervention to regulate eating pace during daily meals. Earinter repurposes the earbud's bone-conduction voice sensor to capture chewing-related vibrations and estimate eating pace as chews per swallow (CPS) for on-device inference. With data collected equally across \modify{sessions in quiet and noisy environments}, Earinter achieves reliable chewing detection (F1 = 0.97) and accurate eating pace estimation (MAE: 0.18 $\pm$ 0.13 chews/min, 3.65 $\pm$ 3.86 chews/swallow), enabling robust tracking for closed-loop use. Guided by Dual Systems Theory and refined through two Wizard-of-Oz pilots, Earinter adopts a user-friendly design for JIT intervention content and delivery policy in daily meals. In a 13-day within-subject field study (N=14), the closed-loop system significantly increased CPS and slowed meal-level eating pace, with statistical signs of carryover on retention-probe days and generally favorable comfort and usability. These effects should be interpreted as changes in eating-pace regulation rather than evidence of weight loss. Our findings highlight how single-modality commodity earables can support practical, theory-driven closed-loop JIT interventions for regulating eating pace in the wild.

cs.HC↗

arXiv2026

From Glance to Scrutiny: Progressive Distortion Reasoning for Fine-Grained Image Quality Assessment

Multi-modal large language models (MLLMs) have demonstrated significant potential in image quality assessment (IQA) by bridging visual perception with descriptive evaluations. However, existing approaches mainly focus on holistic quality prediction, often functioning as black boxes that provide limited insight into where distortions occur and how they affect perceived quality, hindering fine-grained analysis of localized and heterogeneous degradations. We propose GS-IQA, a framework that reformulates IQA as a progressive Where--What--How diagnosis, emulating the human perceptual process from an initial glance to closer scrutiny. Since a severity judgment is meaningful only for a correctly localized and recognized region, we realize this progression through a two-stage reinforcement learning paradigm that respects such dependencies: the glance stage uses a perception-gated reward to establish where degradations lie and what they are, activating severity feedback only once both are correct, while the scrutiny stage introduces online reward-conditioned degradation generation to synthesize hard examples targeted at the model's perceptual bottlenecks, sharpening its discrimination of subtle severity variations. To enable systematic evaluation, we construct Diag-Bench, a region-level IQA benchmark of about 25K curated samples spanning 12 distortion types and five ordinal severity levels. Extensive experiments show that GS-IQA consistently surpasses state-of-the-art methods in distortion localization, recognition, and severity estimation, and that its diagnostic representations transfer effectively to conventional global quality prediction across diverse external benchmarks. Code and data will be released.

cs.CV↗

arXiv2026

Target-Aware Early Stage Ranking

Early Stage Ranking (ESR) in large-scale recommendation systems is dominated by ''user--item decoupling'' Two Tower architectures, which scale efficiently but cannot capture fine-grained, target-aware user--item interactions directly. We propose Target-Aware Early Stage Ranking (TESR), which augments the Two Tower with a Mixture of Attention (MoA) module trained as a request-level sequence modeling over user history. MoA combines (i) Hard Matching Attention (HMA) to capture explicit categorical-ID level overlap signals between user history and candidate item, (ii) target-aware HSTU attention for implicit affinities conditioned on the candidate, and (iii) target dependent and independent cross-attention for symmetric user-item contextualization. On top of this, a Multi-Logit Parameterized Gating (MLPG) head amplifies these signals at scoring time. To keep latency within ESR budgets, we co-design the architecture with FP8 quantization, custom kernels, and a Torch Inductor compilation path. On a production deployment, TESR delivers consistent offline NE wins and online topline gains, and is, to our knowledge, the first deployment of full target-aware attention sequence modeling in an ESR stage at this scale.

cs.LG↗

arXiv2026

Towards Wearable Opportunistic Crowdsensing for Open-Vocabulary Activity Data Collection Through User-Scheduled Trigger-Action Routines

Collecting richly labeled wearable activity data in everyday settings remains difficult because retrospective annotation is costly and often imprecise. Prior data collection apps rely on a labor-intensive self-reporting strategy and primarily treat participants as crowd labelers. We present Pebbl, a feasibility-stage system that incentivizes in-situ labeling through opportunistic crowdsensing. Pebbl lets users author trigger-action recipes on a smartphone and receive just-in-time reminders for beneficial actions when a trigger is detected. In the prototype, triggers are a limited set with four common audio cues, while actions are described in open-vocabulary natural language. Each confirmed execution yields a short sensor window with explicit start/end boundaries and a user-authored action label. We evaluate Pebbl through an expert workshop with wearable Human Activity Recognition (HAR) researchers (N = 6), a within-subject in-lab study (N = 21), and a pilot deployment (N = 8). Experts viewed the approach as lower burden and more ecologically valid than common labeling workflows. In the lab, Pebbl produced reliable execution logs under controlled conditions (recall = 97.30%, precision = 97.15%) and was preferred over comparison workflows on perceived burden and confidence. The pilot deployment shows that the interaction and sensing pipeline can function in free-living use, while surfacing practical constraints such as false triggers and context dependence. Overall, Pebbl represents a step toward a low-burden, distributable collection approach of user-contributed wearable activity data.

cs.HC↗

arXiv2026

RecoWorld: Building Simulated Environments for Agentic Recommender Systems

We present RecoWorld, a blueprint for building simulated environments tailored to agentic recommender systems. Such environments give agents a proper training space where they can learn from errors without impacting real users. RecoWorld distinguishes itself with a dual-view architecture: a simulated user and an agentic recommender engage in multi-turn interactions aimed at maximizing user retention. The user simulator reviews recommended items, updates its mindset, and when sensing potential user disengagement, generates reflective instructions. The agentic recommender adapts its recommendations by incorporating these user instructions and reasoning traces, creating a dynamic feedback loop that actively engages users. This process leverages the exceptional reasoning capabilities of modern LLMs. We explore diverse content representations within the simulator, including text-based, multimodal, and semantic ID modeling, and discuss how multi-turn RL enables the recommender to refine its strategies through iterative interactions. RecoWorld also supports multi-agent simulations, allowing creators to simulate the responses of targeted user populations. It marks an important first step toward recommender systems where users and agents collaboratively shape personalized information streams. We envision new interaction paradigms where "user instructs, recommender responds," jointly optimizing user retention and engagement.

cs.IR↗

arXiv2026

PathRouter: Aligning Rewards with Retrieval Quality in Agentic Graph Retrieval-Augmented Generation

Agentic GraphRAG trains language-model agents to iteratively retrieve and reason over graph-structured evidence, enabling more accurate and context-aware decision-making by efficiently navigating complex information networks. However, outcome-only reinforcement learning suffers from \textit{\textbf{answer-path reward aliasing}}, where correct answers may come from shortcuts rather than useful evidence paths. It also exhibits \textit{\textbf{search-update ambiguity}}, as scalar trajectory-level feedback does not indicate which retrieval actions to adjust. To mitigate these shortcomings, we present PathRouter, a path-aware training framework for agentic GraphRAG. PathRouter jointly evaluates each trajectory along answer correctness and evidence-path overlap, yielding four trajectory categories with differentiated GRPO advantage scaling that suppresses shortcut reinforcement while preserving evidence-seeking behavior. For evidence-poor trajectories, a frozen gold-evidence teacher provides token-level KL guidance on reasoning and search-query tokens, excluding answer tokens to avoid direct response imitation. Experiments on six QA benchmarks across three model sizes show that PathRouter consistently improves answer F1 and evidence-path overlap, achieving average F1 gains of 3.1 on 3B and 4.9 on 7B models compared to a strong baseline.

cs.CL↗

arXiv2026

Do MLLMs Understand Pointing? Benchmarking and Enhancing Referential Reasoning in Egocentric Vision

Egocentric AI agents, such as smart glasses, rely on pointing gestures to resolve referential ambiguities in natural language commands. However, despite advancements in Multimodal Large Language Models (MLLMs), current systems often fail to precisely ground the spatial semantics of pointing. Instead, they rely on spurious correlations with visual proximity or object saliency, a phenomenon we term "Referential Hallucination." To address this gap, we introduce EgoPoint-Bench, a comprehensive question-answering benchmark designed to evaluate and enhance multimodal pointing reasoning in egocentric views. Comprising over 11k high-fidelity simulated and real-world samples, the benchmark spans five evaluation dimensions and three levels of referential complexity. Extensive experiments demonstrate that while state-of-the-art proprietary and open-source models struggle with egocentric pointing, models fine-tuned on our synthetic data achieve significant performance gains and robust sim-to-real generalization. This work highlights the importance of spatially aware supervision and offers a scalable path toward precise egocentric AI assistants. Project page: https://guyyyug.github.io/EgoPoint-Bench/

cs.CV↗

arXiv2026

LLM-Driven Reasoning for Constraint-Aware Feature Selection in Industrial Systems

Feature selection is a crucial step in large-scale industrial machine learning systems, directly affecting model accuracy, efficiency, and maintainability. Traditional feature selection methods rely on labeled data and statistical heuristics, making them difficult to apply in production environments where labeled data are limited and multiple operational constraints must be satisfied. To address this, we propose Model Feature Agent (MoFA), a model-driven framework that performs sequential, reasoning-based feature selection using both semantic and quantitative feature information. MoFA incorporates feature definitions, importance scores, correlations, and metadata (e.g., feature groups or types) into structured prompts and selects features through interpretable, constraint-aware reasoning. We evaluate MoFA in three real-world industrial applications: (1) True Interest and Time-Worthiness Prediction, where it improves accuracy while reducing feature group complexity, (2) Value Model Enhancement, where it discovers high-order interaction terms that yield substantial engagement gains in online experiments, and (3) Notification Behavior Prediction, where it selects compact, high-value feature subsets that improve both model accuracy and inference efficiency. Together, these results demonstrate the practicality and effectiveness of LLM-based reasoning for feature selection in real production systems.

cs.CL↗

arXiv2026

EBPO: Empirical Bayes Shrinkage for Stabilizing Group-Relative Policy Optimization

Reinforcement Learning with Verifiable Rewards (RLVR) has proven effective for enhancing the reasoning capabilities of Large Language Models (LLMs). However, dominant approaches like Group Relative Policy Optimization (GRPO) face critical stability challenges: they suffer from high estimator variance under computational constraints (small group sizes) and vanishing gradient signals in saturated failure regimes where all responses yield identical zero rewards. To address this, we propose Empirical Bayes Policy Optimization (EBPO), a novel framework that regularizes local group-based baselines by borrowing strength from the policy's accumulated global statistics. Instead of estimating baselines in isolation, EBPO employs a shrinkage estimator that dynamically balances local group statistics with a global prior updated via Welford's online algorithm. Theoretically, we demonstrate that EBPO guarantees strictly lower Mean Squared Error (MSE), bounded entropy decay, and non-vanishing penalty signals in failure scenarios compared to GRPO. Empirically, EBPO consistently outperforms GRPO and other established baselines across diverse benchmarks, including AIME and OlympiadBench. Notably, EBPO exhibits superior training stability, achieving high-performance gains even with small group sizes, and benefits significantly from difficulty-stratified curriculum learning.

cs.LG↗

arXiv2025

Mixture-of-Minds: Multi-Agent Reinforcement Learning for Table Understanding

Understanding and reasoning over tables is a critical capability for many real-world applications. Large language models (LLMs) have shown promise on this task, but current approaches remain limited. Fine-tuning based methods strengthen language reasoning; yet they are prone to arithmetic errors and hallucination. In contrast, tool-based methods enable precise table manipulation but rely on rigid schemas and lack semantic understanding. These complementary drawbacks highlight the need for approaches that integrate robust reasoning with reliable table processing. In this work, we propose Mixture-of-Minds, a multi-agent framework that decomposes table reasoning into three specialized roles: planning, coding, and answering. This design enables each agent to focus on a specific aspect of the task while leveraging code execution for precise table manipulation. Building on this workflow, we introduce a self-improvement training framework that employs Monte Carlo Tree Search (MCTS) rollouts to generate pseudo-gold trajectories and optimize agents with reinforcement learning (RL). Extensive experiments show that Mixture-of-Minds delivers substantial gains, reaching 62.13% on TableBench and surpassing OpenAI-o4-mini-high. These results demonstrate the promise of combining structured multi-agent workflows with RL to advance table understanding.

cs.CL↗

arXiv2025

Thought Communication in Multiagent Collaboration

Natural language has long enabled human cooperation, but its lossy, ambiguous, and indirect nature limits the potential of collective intelligence. While machines are not subject to these constraints, most LLM-based multi-agent systems still rely solely on natural language, exchanging tokens or their embeddings. To go beyond language, we introduce a new paradigm, thought communication, which enables agents to interact directly mind-to-mind, akin to telepathy. To uncover these latent thoughts in a principled way, we formalize the process as a general latent variable model, where agent states are generated by an unknown function of underlying thoughts. We prove that, in a nonparametric setting without auxiliary information, both shared and private latent thoughts between any pair of agents can be identified. Moreover, the global structure of thought sharing, including which agents share which thoughts and how these relationships are structured, can also be recovered with theoretical guarantees. Guided by the established theory, we develop a framework that extracts latent thoughts from all agents prior to communication and assigns each agent the relevant thoughts, along with their sharing patterns. This paradigm naturally extends beyond LLMs to all modalities, as most observational data arise from hidden generative processes. Experiments on both synthetic and real-world benchmarks validate the theory and demonstrate the collaborative advantages of thought communication. We hope this work illuminates the potential of leveraging the hidden world, as many challenges remain unsolvable through surface-level observation alone, regardless of compute or data scale.

cs.LG↗

arXiv2025

HyperZero: A Customized End-to-End Auto-Tuning System for Recommendation with Hourly Feedback

Modern recommendation systems can be broadly divided into two key stages: the ranking stage, where the system predicts various user engagements (e.g., click-through rate, like rate, follow rate, watch time), and the value model stage, which aggregates these predictive scores through a function (e.g., a linear combination defined by a weight vector) to measure the value of each content by a single numerical score. Both stages play roughly equally important roles in real industrial systems; however, how to optimize the model weights for the second stage still lacks systematic study. This paper focuses on optimizing the second stage through auto-tuning technology. Although general auto-tuning systems and solutions - both from established production practices and open-source solutions - can address this problem, they typically require weeks or even months to identify a feasible solution. Such prolonged tuning processes are unacceptable in production environments for recommendation systems, as suboptimal value models can severely degrade user experience. An effective auto-tuning solution is required to identify a viable model within 2-3 days, rather than the extended timelines typically associated with existing approaches. In this paper, we introduce a practical auto-tuning system named HyperZero that addresses these time constraints while effectively solving the unique challenges inherent in modern recommendation systems. Moreover, this framework has the potential to be expanded to broader tuning tasks within recommendation systems.

cs.IR↗