RecGPT-V2 Technical Report

Large language models (LLMs) have demonstrated remarkable potential in transforming recommender systems from implicit behavioral pattern matching to explicit intent reasoning. While RecGPT-V1 successfully pioneered this paradigm by integrating LLM-based reasoning into user interest mining and item tag prediction, it suffers from four fundamental limitations: (1) computational inefficiency and cognitive redundancy across multiple reasoning routes; (2) insufficient explanation diversity in fixed-template generation; (3) limited generalization under supervised learning paradigms; and (4) simplistic outcome-focused evaluation that fails to match human standards. To address these challenges, we present RecGPT-V2 with four key innovations. First, a Hierarchical Multi-Agent System restructures intent reasoning through coordinated collaboration, eliminating cognitive duplication while enabling diverse intent coverage. Combined with Hybrid Representation Inference that compresses user-behavior contexts, our framework reduces GPU consumption by 60% and improves exclusive recall from 9.39% to 10.99%. Second, a Meta-Prompting framework dynamically generates contextually adaptive prompts, improving explanation diversity by +7.3%. Third, constrained reinforcement learning mitigates multi-reward conflicts, achieving +24.1% improvement in tag prediction and +13.0% in explanation acceptance. Fourth, an Agent-as-a-Judge framework decomposes assessment into multi-step reasoning, improving human preference alignment. Online A/B tests on Taobao demonstrate significant improvements: +2.98% CTR, +3.71% IPV, +2.19% TV, and +11.46% NER. RecGPT-V2 establishes both the technical feasibility and commercial viability of deploying LLM-powered intent reasoning at scale, bridging the gap between cognitive exploration and industrial utility.

What Affects the Effective Depth of Large Language Models?

The scaling of large language models (LLMs) emphasizes increasing depth, yet performance gains diminish with added layers. Prior work introduces the concept of "effective depth", arguing that deeper models fail to fully utilize their layers for meaningful computation. Building on this, we systematically study how effective depth varies with model scale, training type, and task difficulty. First, we analyze the model behavior of Qwen-2.5 family (1.5B-32B) and find that while the number of effective layers grows with model size, the effective depth ratio remains stable. Besides, comparisons between base and corresponding long-CoT models show no increase in effective depth, suggesting that improved reasoning stems from longer context rather than deeper per-token computation. Furthermore, evaluations across tasks of varying difficulty indicate that models do not dynamically use more layers for harder problems. Our results suggest that current LLMs underuse available depth across scales, training paradigms and tasks of varying difficulties, pointing out research opportunities on increasing the layer utilization rate of LLMs, model pruning, and early exiting. Our code is released at https://github.com/AheadOFpotato/what_affects_effective_depth.

Deep Q-Learning-Based Intelligent Scheduling for ETL Optimization in Heterogeneous Data Environments

This paper addresses the challenges of low scheduling efficiency, unbalanced resource allocation, and poor adaptability in ETL (Extract-Transform-Load) processes under heterogeneous data environments by proposing an intelligent scheduling optimization framework based on deep Q-learning. The framework formalizes the ETL scheduling process as a Markov Decision Process and enables adaptive decision-making by a reinforcement learning agent in high-dimensional state spaces to dynamically optimize task allocation and resource scheduling. The model consists of a state representation module, a feature embedding network, a Q-value estimator, and a reward evaluation mechanism, which collectively consider task dependencies, node load states, and data flow characteristics to derive the optimal scheduling strategy in complex environments. A multi-objective reward function is designed to balance key performance indicators such as average scheduling delay, task completion rate, throughput, and resource utilization. Sensitivity experiments further verify the model's robustness under changes in hyperparameters, environmental dynamics, and data scale. Experimental results show that the proposed deep Q-learning scheduling framework significantly reduces scheduling delay, improves system throughput, and enhances execution stability under multi-source heterogeneous task conditions, demonstrating the strong potential of reinforcement learning in complex data scheduling and resource management, and providing an efficient and scalable optimization strategy for intelligent data pipeline construction.

RecGPT Technical Report

Recommender systems are among the most impactful applications of artificial intelligence, serving as critical infrastructure connecting users, merchants, and platforms. However, most current industrial systems remain heavily reliant on historical co-occurrence patterns and log-fitting objectives, i.e., optimizing for past user interactions without explicitly modeling user intent. This log-fitting approach often leads to overfitting to narrow historical preferences, failing to capture users' evolving and latent interests. As a result, it reinforces filter bubbles and long-tail phenomena, ultimately harming user experience and threatening the sustainability of the whole recommendation ecosystem. To address these challenges, we rethink the overall design paradigm of recommender systems and propose RecGPT, a next-generation framework that places user intent at the center of the recommendation pipeline. By integrating large language models (LLMs) into key stages of user interest mining, item retrieval, and explanation generation, RecGPT transforms log-fitting recommendation into an intent-centric process. To effectively align general-purpose LLMs to the above domain-specific recommendation tasks at scale, RecGPT incorporates a multi-stage training paradigm, which integrates reasoning-enhanced pre-alignment and self-training evolution, guided by a Human-LLM cooperative judge system. Currently, RecGPT has been fully deployed on the Taobao App. Online experiments demonstrate that RecGPT achieves consistent performance gains across stakeholders: users benefit from increased content diversity and satisfaction, merchants and the platform gain greater exposure and conversions. These comprehensive improvement results across all stakeholders validates that LLM-driven, intent-centric design can foster a more sustainable and mutually beneficial recommendation ecosystem.

EXAONE Path 2.0: Pathology Foundation Model with End-to-End Supervision

In digital pathology, whole-slide images (WSIs) are often difficult to handle due to their gigapixel scale, so most approaches train patch encoders via self-supervised learning (SSL) and then aggregate the patch-level embeddings via multiple instance learning (MIL) or slide encoders for downstream tasks. However, patch-level SSL may overlook complex domain-specific features that are essential for biomarker prediction, such as mutation status and molecular characteristics, as SSL methods rely only on basic augmentations selected for natural image domains on small patch-level area. Moreover, SSL methods remain less data efficient than fully supervised approaches, requiring extensive computational resources and datasets to achieve competitive performance. To address these limitations, we present EXAONE Path 2.0, a pathology foundation model that learns patch-level representations under direct slide-level supervision. Using only 37k WSIs for training, EXAONE Path 2.0 achieves state-of-the-art average performance across 10 biomarker prediction tasks, demonstrating remarkable data efficiency.

From Local Patterns to Global Understanding: Cross-Stock Trend Integration for Enhanced Predictive Modeling

Stock price prediction is a critical area of financial forecasting, traditionally approached by training models using the historical price data of individual stocks. While these models effectively capture single-stock patterns, they fail to leverage potential correlations among stock trends, which could improve predictive performance. Current single-stock learning methods are thus limited in their ability to provide a broader understanding of price dynamics across multiple stocks. To address this, we propose a novel method that merges local patterns into a global understanding through cross-stock pattern integration. Our strategy is inspired by Federated Learning (FL), a paradigm designed for decentralized model training. FL enables collaborative learning across distributed datasets without sharing raw data, facilitating the aggregation of global insights while preserving data privacy. In our adaptation, we train models on individual stock data and iteratively merge them to create a unified global model. This global model is subsequently fine-tuned on specific stock data to retain local relevance. The proposed strategy enables parallel training of individual stock models, facilitating efficient utilization of computational resources and reducing overall training time. We conducted extensive experiments to evaluate the proposed method, demonstrating that it outperforms benchmark models and enhances the predictive capabilities of state-of-the-art approaches. Our results highlight the efficacy of Cross-Stock Trend Integration (CSTI) in advancing stock price prediction, offering a robust alternative to traditional single-stock learning methodologies.

PHYBench: Holistic Evaluation of Physical Perception and Reasoning in Large Language Models

We introduce PHYBench, a novel, high-quality benchmark designed for evaluating reasoning capabilities of large language models (LLMs) in physical contexts. PHYBench consists of 500 meticulously curated physics problems based on real-world physical scenarios, designed to assess the ability of models to understand and reason about realistic physical processes. Covering mechanics, electromagnetism, thermodynamics, optics, modern physics, and advanced physics, the benchmark spans difficulty levels from high school exercises to undergraduate problems and Physics Olympiad challenges. Additionally, we propose the Expression Edit Distance (EED) Score, a novel evaluation metric based on the edit distance between mathematical expressions, which effectively captures differences in model reasoning processes and results beyond traditional binary scoring methods. We evaluate various LLMs on PHYBench and compare their performance with human experts. Our results reveal that even state-of-the-art reasoning models significantly lag behind human experts, highlighting their limitations and the need for improvement in complex physical reasoning scenarios. Our benchmark results and dataset are publicly available at https://phybench-official.github.io/phybench-demo/.

Tin-Tin: Towards Tiny Learning on Tiny Devices with Integer-based Neural Network Training

Recent advancements in machine learning (ML) have enabled its deployment on resource-constrained edge devices, fostering innovative applications such as intelligent environmental sensing. However, these devices, particularly microcontrollers (MCUs), face substantial challenges due to limited memory, computing capabilities, and the absence of dedicated floating-point units (FPUs). These constraints hinder the deployment of complex ML models, especially those requiring lifelong learning capabilities. To address these challenges, we propose Tin-Tin, an integer-based on-device training framework designed specifically for low-power MCUs. Tin-Tin introduces novel integer rescaling techniques to efficiently manage dynamic ranges and facilitate efficient weight updates using integer data types. Unlike existing methods optimized for devices with FPUs, GPUs, or FPGAs, Tin-Tin addresses the unique demands of tiny MCUs, prioritizing energy efficiency and optimized memory utilization. We validate the effectiveness of Tin-Tin through end-to-end application examples on real-world tiny devices, demonstrating its potential to support energy-efficient and sustainable ML applications on edge platforms.

Training Large Language Models to be Better Rule Followers

Large language models (LLMs) have shown impressive performance across a wide range of tasks. However, they often exhibit unexpected failures in seemingly straightforward tasks, suggesting a reliance on case-based reasoning rather than rule-based reasoning. While the vast training corpus of LLMs contains numerous textual "rules", current training methods fail to leverage these rules effectively. Crucially, the relationships between these "rules" and their corresponding "instances" are not explicitly modeled. As a result, while LLMs can often recall rules with ease, they fail to apply these rules strictly and consistently in relevant reasoning scenarios. In this paper, we investigate the rule-following capabilities of LLMs and propose Meta Rule-Following Fine-Tuning (Meta-RFFT) to enhance the cross-task transferability of rule-following abilities. We first construct a dataset of 88 tasks requiring following rules, encompassing diverse reasoning domains. We demonstrate through extensive experiments that models trained on large-scale rule-following tasks are better rule followers, outperforming the baselines in both downstream fine-tuning and few-shot prompting scenarios. This highlights the cross-task transferability of models with the aid of Meta-RFFT. Furthermore, we examine the influence of factors such as dataset size, rule formulation, and in-context learning.

RedStar: Does Scaling Long-CoT Data Unlock Better Slow-Reasoning Systems?

Can scaling transform reasoning? In this work, we explore the untapped potential of scaling Long Chain-of-Thought (Long-CoT) data to 1000k samples, pioneering the development of a slow-thinking model, RedStar. Through extensive experiments with various LLMs and different sizes, we uncover the ingredients for specialization and scale for Long-CoT training. Surprisingly, even smaller models show significant performance gains with limited data, revealing the sample efficiency of Long-CoT and the critical role of sample difficulty in the learning process. Our findings demonstrate that Long-CoT reasoning can be effectively triggered with just a few thousand examples, while larger models achieve unparalleled improvements. We also introduce reinforcement learning (RL)-scale training as a promising direction for advancing slow-thinking systems. RedStar shines across domains: on the MATH-Hard benchmark, RedStar-code-math boosts performance from 66.2\% to 81.6\%, and on the USA Math Olympiad (AIME), it solves 46.7\% of problems using only 21k mixed-code-math datasets. In multimodal tasks like GeoQA and MathVista-GEO, RedStar-Geo achieves competitive results with minimal Long-CoT data, outperforming other slow-thinking systems like QvQ-Preview. Compared to QwQ, RedStar strikes the perfect balance between reasoning and generalizability. Our work highlights that, with careful tuning, scaling Long-CoT can unlock extraordinary reasoning capabilities-even with limited dataset and set a new standard for slow-thinking models across diverse challenges. Our data and models are released at https://huggingface.co/RedStar-Reasoning.