reward modeling
26 articles about reward modeling in AI news
Robots Learn Self-Supervised Progress Tracking via Reward Modeling Survey
Survey unifies progress reward modeling for robots to self-assess advancement, stagnation, or regression during tasks, replacing binary success signals.
U.K. Retail Loyalty Enters AI Era as M&S
Marks & Spencer, Tesco, and Boots are implementing AI to analyze customer data and deliver hyper-personalized rewards and offers within their loyalty programs. This marks a strategic shift from one-size-fits-all schemes to predictive, individualized engagement to boost retention and spending.
LLMs Learn to Switch Reasoning Effort at Inference Time
@rasbt explains how LLMs switch reasoning effort using inference-time methods and training, potentially cutting token usage by 30–50% on simple queries.
Visual-SDPO: Self-Distillation Fixes Code-Generated Visual Defects by +10 Points
Visual-SDPO uses visual-feedback self-distillation to improve code-generated visual artifacts by >10 points on ChartMimic, Design2Code, and AeSlides, with no added inference cost.
Guerlain Launches First Paid Influencer Campaign After Viral TikTok
Guerlain reports the Vanille Planifolia extrait became its #1 best-selling product for five months after organic TikTok videos, leading to the brand’s first paid influencer campaign. Sales tripled despite the $660 price, and the fragrance sold out multiple times.
New Research Models 'Exploration Saturation' in Recommender Systems
A research paper analyzes 'exploration saturation'—the point where more diverse recommendations hurt user utility. Findings show this saturation point is user-dependent, challenging the standard practice of applying uniform fairness or novelty pressure across all users.
Rapid Interest Shifts in Recommender Systems: A Case Study on Instagram Reels
A personal experiment demonstrates the remarkable speed at which Instagram's Reels recommendation system detects and responds to changes in user engagement patterns, highlighting the real-time adaptability of modern algorithms.
New Research Proposes Unified LLM Framework for Need-Driven Service
A new arXiv paper introduces a large language model framework that unifies living need prediction and service recommendation for local life services. It uses behavioral clustering to filter noise and a curriculum learning + RL strategy to navigate complex decision paths. Experiments show it significantly improves both need prediction and recommendation accuracy.
SPPO: Sequence-Level PPO Cuts RL Training Time 5.9x for Math Reasoning
Researchers introduced SPPO, a sequence-level PPO algorithm that reformulates reasoning as a contextual bandit. It achieves a 5.9x speedup over GRPO while matching performance on AIME, AMC, and MATH benchmarks at 1.5B and 7B scales.
Claude AI Prompts Claim to Build Hedge Fund-Level Trading Strategies
A prompt collection claims to enable Claude to build and backtest hedge fund-level trading strategies. The prompts aim to automate quantitative analysis tasks typically performed by high-paid analysts.
Coupang Eats Secures Patent for Budget-Based Food Recommendation System
Coupang Eats has been granted a patent for a food recommendation engine that factors in a user's defined budget. This system aims to provide more relevant suggestions than basic price filters by integrating budget as a core ranking signal. It represents a strategic move to enhance user experience and conversion in the competitive delivery market.
ASI-Evolve: This AI Designs Better AI Than Humans Can — 105 New Architectures, Zero Human Guidance
Researchers built an AI that runs the entire research cycle on its own — reading papers, designing experiments, running them, and learning from results. It discovered 105 architectures that beat human-designed models, and invented new learning algorithms. Open-sourced.
Goal-Aligned Recommendation Systems: Lessons from Return-Aligned Decision Transformer
The article discusses Return-Aligned Decision Transformer (RADT), a method that aligns recommender systems with long-term business returns. It addresses the common problem where models ignore target signals, offering a framework for transaction-driven recommendations.
arXiv Paper Proposes 'Connections' Word Game as New Benchmark for AI Agent Social Intelligence
A new arXiv preprint introduces the improvisational word game 'Connections' as a benchmark for evaluating social intelligence in AI agents. It requires agents to gauge the cognitive states of others, testing collaborative reasoning beyond individual knowledge retrieval.
Harvard Business Review Presents AI Agent Governance Framework: Job Descriptions, Limits, and Managers Required
Harvard Business Review argues AI agents must be managed like employees with defined roles, permissions, and audit trails, proposing a four-layer safety framework and an 'autonomy ladder' for gradual deployment.
New Research Proposes Lightweight Framework for Adapting LLMs to Complex Service Domains
A new arXiv paper introduces a three-part framework to efficiently adapt LLMs for technical service agents. It addresses latent decision logic, response ambiguity, and high training costs, validated on cloud service tasks. This matters for any domain needing robust, specialized AI agents.
Multi-Agent Reinforcement Learning for Dynamic Pricing: A Comparative Study of MAPPO and MADDPG
A new arXiv paper benchmarks multi-agent RL algorithms for competitive dynamic pricing. MAPPO achieved the highest, most stable profits, while MADDPG delivered the fairest outcomes. This offers a scalable alternative to independent learning for retail price optimization.
LLM-Driven Motivation-Aware Multimodal Recommendation (LMMRec): A New Framework for Understanding User Intent
Researchers propose LMMRec, a model-agnostic framework using LLMs to extract fine-grained user and item motivations from text. It aligns textual and interaction-based motivations, achieving up to 4.98% performance gains on three datasets.
Hermès Faces Questions as Birkin and Kelly Resale Market Softens
The Business of Fashion reports a softening resale market for Hermès's iconic Birkin and Kelly bags, posing strategic questions for the luxury powerhouse. This signals a potential shift in the ultra-luxury asset class.
Beyond One-Size-Fits-All AI: New Method Aligns Language Models with Diverse Human Preferences
Researchers have developed Personalized GRPO, a novel reinforcement learning framework that enables large language models to align with heterogeneous human preferences rather than optimizing for a single global objective. The approach addresses systematic bias toward dominant preferences in current alignment methods.
AI Researchers Solve Critical LLM Confidence Problem with Novel Decoupling Technique
Researchers have identified and solved a fundamental conflict in how large language models learn reasoning versus confidence calibration. Their new DCPO framework preserves reasoning accuracy while dramatically reducing overconfidence in incorrect answers, addressing a major reliability concern for AI deployment.
Beyond Sequence Generation: The Emergence of Agentic Reinforcement Learning for LLMs
A new survey paper argues that LLM reinforcement learning must evolve beyond narrow sequence generation to embrace true agentic capabilities. The research introduces a comprehensive taxonomy for agentic RL, mapping environments, benchmarks, and frameworks shaping this emerging field.
VAST's $50M Funding Signals 3D AI Revolution: From Foundation Models to World Simulation
AI startup VAST has secured $50 million in Series A funding while advancing its 3D foundation models that are setting new industry standards. The company is preparing to launch its first world model, positioning itself at the forefront of spatial AI development.
Beyond A/B Testing: How Constraint-Aware Generative AI is Revolutionizing E-commerce Ranking
New research introduces a unified neural framework for generative re-ranking that optimizes for multiple business objectives (like revenue and engagement) while respecting real-time constraints. This enables luxury retailers to dynamically personalize product feeds, balancing commercial goals with brand experience.
ByteDance's Molecular AI Breakthrough: Stabilizing Complex Reasoning with Chemical Bond Principles
ByteDance researchers have developed MOLE-SYN, a novel AI approach that maps molecular bond dynamics to stabilize long-chain reasoning in language models. This breakthrough addresses the 'cold-start' problem in multi-step AI reasoning and enhances reinforcement learning stability.
Living Architecture: AI-Designed Cyanobacteria Concrete That Repairs Itself and Captures Carbon
Researchers have developed a revolutionary living building material using cyanobacteria that captures atmospheric CO₂ and self-reinforces over time. This bio-concrete, validated by 400+ days of laboratory data, represents a paradigm shift toward regenerative construction.