Research & Analysis

In-depth research papers, case studies, surveys, and academic insights on finance and technology

Articles

436

Updated

May 7, 2026

Why Global LLM Leaderboards Are Misleading: Small Portfolios for Heterogeneous Supervised ML

Section 1 – What happened? A recent study has questioned the accuracy of global leaderboards ranking Large Language Models (LLMs) in open-ended tasks…

Research & Analysis

Why Global LLM Leaderboards Are Misleading: Small Portfolios for Heterogeneous Supervised ML

Top Stories

Verifier-Backed Hard Problem Generation for Mathematical Reasoning

UniPool: A Globally Shared Expert Pool for Mixture-of-Experts

BAMI: Training-Free Bias Mitigation in GUI Grounding

ActCam: Zero-Shot Joint Camera and 3D Motion Control for Video Generation

More Stories

Why Global LLM Leaderboards Are Misleading: Small Portfolios for Heterogeneous Supervised ML

Verifier-Backed Hard Problem Generation for Mathematical Reasoning

UniPool: A Globally Shared Expert Pool for Mixture-of-Experts

BAMI: Training-Free Bias Mitigation in GUI Grounding

ActCam: Zero-Shot Joint Camera and 3D Motion Control for Video Generation

Taming Outlier Tokens in Diffusion Transformers

Sharp Capacity Thresholds in Linear Associative Memory: From Winner-Take-All to Listwise Retrieval

LongSeeker: Elastic Context Orchestration for Long-Horizon Search Agents

Grokability in five inequalities

Almost-Orthogonality in Lp Spaces: A Case Study with Grok

Taming Outlier Tokens in Diffusion Transformers

Sharp Capacity Thresholds in Linear Associative Memory: From Winner-Take-All to Listwise Retrieval

LongSeeker: Elastic Context Orchestration for Long-Horizon Search Agents

Grokability in five inequalities

Almost-Orthogonality in Lp Spaces: A Case Study with Grok

Safety and accuracy follow different scaling laws in clinical large language models

Redefining AI Red Teaming in the Agentic Era: From Weeks to Hours

OpenSeeker-v2: Pushing the Limits of Search Agents with Informative and High-Difficulty Trajectories

Large-Scale High-Quality 3D Gaussian Head Reconstruction from Multi-View Captures

A Closed-Form Adaptive-Landmark Kernel for Certified Point-Cloud and Graph Classification

Safety and accuracy follow different scaling laws in clinical large language models

Redefining AI Red Teaming in the Agentic Era: From Weeks to Hours

OpenSeeker-v2: Pushing the Limits of Search Agents with Informative and High-Difficulty Trajectories

Large-Scale High-Quality 3D Gaussian Head Reconstruction from Multi-View Captures

A Closed-Form Adaptive-Landmark Kernel for Certified Point-Cloud and Graph Classification

Unsupervised Machine Learning for Detecting Structural Anomalies in European Regional Statistics

Standing on the Shoulders of Giants: Stabilized Knowledge Distillation for Cross--Language Code Clone Detection

SpecKV: Adaptive Speculative Decoding with Compression-Aware Gamma Selection

Multi-fidelity surrogates for mechanics of composites: from co-kriging to multi-fidelity neural networks

Enhancing RL Generalizability in Robotics through SHAP Analysis of Algorithms and Hyperparameters

Unsupervised Machine Learning for Detecting Structural Anomalies in European Regional Statistics

Standing on the Shoulders of Giants: Stabilized Knowledge Distillation for Cross--Language Code Clone Detection

SpecKV: Adaptive Speculative Decoding with Compression-Aware Gamma Selection

Multi-fidelity surrogates for mechanics of composites: from co-kriging to multi-fidelity neural networks

Enhancing RL Generalizability in Robotics through SHAP Analysis of Algorithms and Hyperparameters

Synthetic Computers at Scale for Long-Horizon Productivity Simulation

LLM as Clinical Graph Structure Refiner: Enhancing Representation Learning in EEG Seizure Diagnosis

Exploration Hacking: Can LLMs Learn to Resist RL Training?

Computing Equilibrium beyond Unilateral Deviation

An adaptive wavelet-based PINN for problems with localized high-magnitude source

Synthetic Computers at Scale for Long-Horizon Productivity Simulation

LLM as Clinical Graph Structure Refiner: Enhancing Representation Learning in EEG Seizure Diagnosis

Exploration Hacking: Can LLMs Learn to Resist RL Training?

Computing Equilibrium beyond Unilateral Deviation

An adaptive wavelet-based PINN for problems with localized high-magnitude source

Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models

On the Learning Curves of Revenue Maximization

Learning Over-Relaxation Policies for ADMM with Convergence Guarantees

Hyper Input Convex Neural Networks for Shape Constrained Learning and Optimal Transport

A Note on How to Remove the $\ln\ln T$ Term from the Squint Bound

Turning the TIDE: Cross-Architecture Distillation for Diffusion Large Language Models

On the Learning Curves of Revenue Maximization

Learning Over-Relaxation Policies for ADMM with Convergence Guarantees

Hyper Input Convex Neural Networks for Shape Constrained Learning and Optimal Transport

A Note on How to Remove the $\ln\ln T$ Term from the Squint Bound

Toward a Functional Geometric Algebra for Natural Language Semantics

Teacher Forcing as Generalized Bayes: Optimization Geometry Mismatch in Switching Surrogates for Chaotic Dynamics

Recursive Multi-Agent Systems

How Fast Should a Model Commit to Supervision? Training Reasoning Models on the Tsallis Loss Continuum

Carbon-Taxed Transformers: A Green Compression Pipeline for Overgrown Language Models

Toward a Functional Geometric Algebra for Natural Language Semantics

Teacher Forcing as Generalized Bayes: Optimization Geometry Mismatch in Switching Surrogates for Chaotic Dynamics

Recursive Multi-Agent Systems

How Fast Should a Model Commit to Supervision? Training Reasoning Models on the Tsallis Loss Continuum

Carbon-Taxed Transformers: A Green Compression Pipeline for Overgrown Language Models

The Optimal Sample Complexity of Multiclass and List Learning

SpecRLBench: A Benchmark for Generalization in Specification-Guided Reinforcement Learning

Personalized Worked Example Generation from Student Code Submissions using Pattern-based Knowledge Components

Learning to Think from Multiple Thinkers

Conflict-Aware Harmonized Rotational Gradient for Multiscale Kinetic Regimes

The Optimal Sample Complexity of Multiclass and List Learning

SpecRLBench: A Benchmark for Generalization in Specification-Guided Reinforcement Learning