Tagged articles

Test-Time Scaling

22 articles · Page 1 of 1

Oct 26, 2026 · Artificial Intelligence

27B AREX-2 Beats 1.6T Models: Long-Horizon Reflection Unlocks Trainable RSI Path

The AREX-2 model, a 27B parameter model trained with long-horizon reflection on machine learning and algorithmic programming tasks, outperforms much larger models on benchmarks like MLE-bench and BrowseComp, demonstrating that reflective self-improvement can transfer across domains and provide a verifiable path to recursive self-improvement (RSI).

AI ResearchAREX-2BrowseComp

0 likes · 12 min read

27B AREX-2 Beats 1.6T Models: Long-Horizon Reflection Unlocks Trainable RSI Path

Machine Heart

Oct 25, 2026 · Artificial Intelligence

ParallelWorld: Multi-Step Lookahead Planning Boosts Embodied Agent Success from 56% to 83%

ParallelWorld introduces test-time scaling for embodied agents by simulating multiple future action paths before execution, improving task success rates from 56% to 83% on benchmarks like ESI-Bench and LIBERO-Pro, with adaptive search width balancing performance and compute cost.

ESI-BenchLIBERO-ProParallelWorld

0 likes · 11 min read

ParallelWorld: Multi-Step Lookahead Planning Boosts Embodied Agent Success from 56% to 83%

Machine Learning Algorithms & Natural Language Processing

Oct 5, 2026 · Artificial Intelligence

Diffusion Reward Models: Learning Human Preference Distributions Beyond Scalar Scores

This paper introduces Diffusion Reward Models (DRM) that replace the scalar value head with a Diffusion Transformer to model full reward distributions, capturing multimodal human disagreement, enabling uncertainty-aware decisions, test-time scaling via repeated reward sampling, and improving downstream RLHF performance.

Distribution ModelingHuman Preference LearningMultimodal Distributions

0 likes · 20 min read

Diffusion Reward Models: Learning Human Preference Distributions Beyond Scalar Scores

Machine Heart

Sep 16, 2026 · Artificial Intelligence

Harness Evolution vs. Test-Time Scaling: Simple Retries Outperform Complex Self-Improvement

A study from AI2 and University of Washington finds that complex Harness Evolution for AI agents fails to consistently outperform simple test-time scaling methods like parallel sampling under equal compute budgets, and improvements rarely transfer to unseen tasks, questioning whether observed gains stem from genuine self-improvement or just extra attempts.

AI AgentsHarness EvolutionParallel Sampling

0 likes · 14 min read

Harness Evolution vs. Test-Time Scaling: Simple Retries Outperform Complex Self-Improvement

Data Party THU

Sep 15, 2026 · Artificial Intelligence

RTWI: Reliable Multimodal Reasoning by Filtering Noisy Thinking

Sichuan University researchers propose RTWI, a plug-and-play test-time scaling framework that models reasoning reliability to filter noisy thinking in multimodal chain-of-thought, achieving higher accuracy and efficiency on benchmarks like Qwen3-VL and DeepEyes.

DeepEyesICML 2026Multimodal Reasoning

0 likes · 13 min read

RTWI: Reliable Multimodal Reasoning by Filtering Noisy Thinking

Xiaomi Tech

Sep 5, 2026 · Artificial Intelligence

Xiaomi-TabLDM: One Model for All Tabular Tasks via Synthetic Pretraining & Test-Time Scaling

Xiaomi releases Xiaomi-TabLDM, a foundation model for tabular data that uses large-scale synthetic pretraining, efficient model scaling with dual-stream feature groups and sparse MoE, and test-time scaling to achieve top-tier performance on four public benchmarks and real-world industrial tasks without per-dataset retraining.

Mixture of ExpertsTest-Time ScalingXiaomi-TabLDM

0 likes · 9 min read

Xiaomi-TabLDM: One Model for All Tabular Tasks via Synthetic Pretraining & Test-Time Scaling

Data Party THU

Jul 22, 2026 · Artificial Intelligence

Viewing LLM Inference as Statistical Decision-Making: Theory, Algorithms, and Experiments

This article surveys a UC Berkeley PhD thesis that reframes large language model inference as a statistical decision process, covering test‑time scaling, diffusion‑model decoding, and statistical watermarking, and presents new algorithms with theoretical guarantees and empirical results.

LLMSEALTest-Time Scaling

0 likes · 15 min read

Viewing LLM Inference as Statistical Decision-Making: Theory, Algorithms, and Experiments

AI2ML AI to Machine Learning

Jul 13, 2026 · Artificial Intelligence

From QA to Task‑Oriented Agents: Recent Trends in Large Language Models

The article surveys the latest advances in large language model agents, covering multi‑agent collaboration, long‑horizon planning, self‑evolution, trust and safety, test‑time scaling techniques, new foundation and multimodal models, open‑source and closed‑source breakthroughs, world‑model integration, and emerging vertical applications.

Foundation ModelsKnowledge GraphsLLM Agents

0 likes · 12 min read

From QA to Task‑Oriented Agents: Recent Trends in Large Language Models

PaperAgent

May 20, 2026 · Artificial Intelligence

AutoTTS Shows How AI Agents Can Outperform Human‑Designed Test‑Time Scaling Strategies

The paper “LLMs Improving LLMs” introduces AutoTTS, an environment where a Claude‑based explorer agent automatically searches test‑time scaling policies, achieving up to 69.5% token savings and superior accuracy on unseen models, all for $39.9 and 160 minutes without any LLM calls during evaluation.

AutoTTSClaudeLLM Agents

0 likes · 7 min read

AutoTTS Shows How AI Agents Can Outperform Human‑Designed Test‑Time Scaling Strategies

Machine Heart

May 11, 2026 · Artificial Intelligence

How PRISM Enables Efficient Test‑Time Scaling for Discrete Diffusion Language Models

The article analyzes how the PRISM framework redesigns test‑time scaling for discrete diffusion language models by replacing costly Best‑of‑N sampling with a three‑stage hierarchical search, local branching via partial remasking, and self‑verified feedback, achieving large accuracy gains on math and code benchmarks while cutting inference compute by up to four‑fold.

Discrete DiffusionHierarchical SearchSelf‑Verification

0 likes · 11 min read

How PRISM Enables Efficient Test‑Time Scaling for Discrete Diffusion Language Models

Machine Heart

Apr 25, 2026 · Artificial Intelligence

Open‑Source Models Dominate 21 Scientific Discovery Tasks with SimpleTES

The SimpleTES framework decomposes trial‑and‑error into three scalable dimensions—Concurrency, Length, and Candidates—enabling test‑time scaling that lets open‑source models outperform closed‑source rivals across 21 diverse scientific benchmarks, from LASSO regression to quantum circuit compilation.

AI for scienceOpen‑source ModelsScientific Discovery

0 likes · 13 min read

Open‑Source Models Dominate 21 Scientific Discovery Tasks with SimpleTES

Machine Heart

Apr 25, 2026 · Artificial Intelligence

Can Multi-Model Co-Evolution Shatter the Single-Model Ceiling? Squeeze Evolve Achieves Validator-Free SOTA Inference

The paper introduces Squeeze Evolve, a validator‑free multi‑model evolutionary framework that orchestrates diverse large language models to break the performance ceiling of any single model, delivering up to 23‑point accuracy improvements and 1.4‑3.3× cost reductions across math, vision, and scientific benchmarks.

AI ResearchLarge Language ModelsSqueeze Evolve

0 likes · 8 min read

Can Multi-Model Co-Evolution Shatter the Single-Model Ceiling? Squeeze Evolve Achieves Validator-Free SOTA Inference

Old Zhang's AI Learning

Jan 27, 2026 · Artificial Intelligence

Qwen3‑Max‑Thinking Boosts Performance with Test‑Time Scaling—Why It Still Isn’t Open‑Source

Alibaba’s new Qwen3‑Max‑Thinking model adds inference‑time scaling and adaptive tool use, delivering large gains on math, coding, and agent benchmarks while remaining closed‑source, and it offers drop‑in OpenAI‑compatible API access at the cost of higher latency and token usage.

AI benchmarkAdaptive Tool UseOpenAI API Compatibility

0 likes · 7 min read

Qwen3‑Max‑Thinking Boosts Performance with Test‑Time Scaling—Why It Still Isn’t Open‑Source

Data Party THU

Oct 29, 2025 · Artificial Intelligence

Can Test-Time Scaling Unlock More Reliable Vision‑Language‑Action Robots?

The paper introduces RoboMonkey, a framework that applies a generate‑and‑verify paradigm and test‑time scaling to Vision‑Language‑Action models, showing that increasing sampling and verification at inference dramatically reduces action error across multiple VLA architectures, and presents scalable verifier training, synthetic data augmentation, and efficient deployment strategies.

AI ResearchAction VerificationRoboMonkey

0 likes · 8 min read

Can Test-Time Scaling Unlock More Reliable Vision‑Language‑Action Robots?

vivo Internet Technology

Aug 25, 2025 · Artificial Intelligence

How DiMo-GUI Boosts Multimodal LLMs for GUI Grounding Without Training

DiMo-GUI is a plug‑and‑play framework that dramatically improves multimodal large language models' ability to locate GUI elements by using a hierarchical dynamic visual reasoning loop and modality‑aware optimization, achieving up to double the performance on high‑resolution GUI benchmarks without any additional training data.

GUI groundingTest-Time Scalingdynamic visual reasoning

0 likes · 7 min read

How DiMo-GUI Boosts Multimodal LLMs for GUI Grounding Without Training

Software Engineering 3.0 Era

Aug 2, 2025 · Artificial Intelligence

How ByteDance’s TRAE Agent Redefines AI-Powered Software Engineering

ByteDance’s TRAE Agent achieves a record 75.20% success on the SWE‑bench benchmark by bridging the “complexity gap” between function‑level and repository‑level tasks through a three‑stage pipeline—patch generation, pruning, and selection—augmented with ensemble reasoning, multi‑model integration, and a novel test‑time scaling mechanism.

AI AgentsSWE-BenchTest-Time Scaling

0 likes · 14 min read

How ByteDance’s TRAE Agent Redefines AI-Powered Software Engineering

Kuaishou Large Model

Jul 3, 2025 · Artificial Intelligence

How EvoSearch Boosts Image & Video Generation with Test‑Time Evolutionary Search

The EvoSearch method introduced by HKUST and Kuaishou’s KuaLing team leverages test‑time scaling to dramatically improve diffusion‑based image and video generation without training, using evolutionary search along the denoising trajectory, achieving state‑of‑the‑art results on SD2.1, Flux‑1‑dev and other models.

Evolutionary SearchTest-Time Scalingdiffusion models

0 likes · 8 min read

How EvoSearch Boosts Image & Video Generation with Test‑Time Evolutionary Search

Kuaishou Tech

Jul 2, 2025 · Artificial Intelligence

How EvoSearch Supercharges Image and Video Generation with Test‑Time Evolutionary Search

EvoSearch, a test‑time evolutionary search method, dramatically improves image and video generation by increasing inference compute without extra training, outperforming existing scaling techniques on diffusion and flow models while maintaining robustness and diversity across multiple benchmarks.

AI ResearchEvolutionary SearchTest-Time Scaling

0 likes · 8 min read

How EvoSearch Supercharges Image and Video Generation with Test‑Time Evolutionary Search

Fighter's World

Jun 28, 2025 · Artificial Intelligence

What Is the Generator‑Verifier Gap and Why It Matters for LLM Reasoning

The article explains the Generator‑Verifier Gap (GVG)—the asymmetry where verifying a solution is far cheaper than generating it—covers its origin, its impact on test‑time scaling for large language models, reinforcement‑learning approaches, and how the concept can shape agent architectures and AI product strategy.

Agent ArchitectureGenerator-Verifier GapLLM

0 likes · 21 min read

What Is the Generator‑Verifier Gap and Why It Matters for LLM Reasoning

AI Frontier Lectures

Apr 4, 2025 · Artificial Intelligence

Why Test‑Time Scaling Is Revolutionizing LLM Reasoning in 2025

This article surveys the latest research on large language model reasoning, highlighting test‑time scaling methods, chain‑of‑thought variants, and novel inference‑time techniques that boost performance while exposing trade‑offs, costs, and future directions for AI developers.

AILLMTest-Time Scaling

0 likes · 26 min read

Why Test‑Time Scaling Is Revolutionizing LLM Reasoning in 2025

Architect

Feb 19, 2025 · Artificial Intelligence

Does Scaling Law Still Hold for Grok 3? A Deep Dive into LLM Training Economics

The article critically examines whether the pre‑training Scaling Law still applies to Grok 3, compares its compute usage and model size with DeepSeek and OpenAI models, evaluates the cost‑effectiveness of pre‑training, RL and test‑time scaling, and explores how these insights shape future large‑language‑model development strategies.

Grok 3Large Language ModelsModel Efficiency

0 likes · 11 min read

Does Scaling Law Still Hold for Grok 3? A Deep Dive into LLM Training Economics

Baobao Algorithm Notes

Sep 29, 2024 · Artificial Intelligence

Decoding OpenAI o1: Test‑Time Scaling, PRM Search & Inference Strategies

This article analyses the training tricks behind OpenAI's o1 model, explaining test/inference‑time scaling laws, post‑training techniques, process‑supervised reward models (PRM), various inference‑time search methods, data‑collection pipelines, and the trade‑offs between allocating compute to pre‑training versus inference.

LLM inferenceOpenAI o1Post-Training

0 likes · 34 min read