A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Lodewijk Gelauff, Mohak Goyal, Bhargav Dindukurthi, et al.
This paper uses an LLM alongside human annotators to rate contribution quality in online deliberations, finding the model competitive with individual humans and demonstrating its utility for evaluating nudges.
Jordan Pollack, Alan Blair
This paper demonstrates that simple hillclimbing with relative fitness can train a neural network to play backgammon at a competitive level, challenging the necessity of temporal difference learning.
Iulian Vlad Serban, Alessandro Sordoni, Yoshua Bengio, et al.
Extends hierarchical recurrent encoder-decoder networks to open-domain dialogue, showing competitiveness with n-gram models and improvement via bootstrapping from QA pairs and pretrained embeddings.
Seyedali Mirjalili, Seyed Mohammad Mirjalili, Seyed Mohammad Mirjalili, et al.
Proposes Grey Wolf Optimizer (GWO), a meta-heuristic inspired by grey wolf leadership hierarchy and hunting behavior, achieving competitive results on benchmark functions and engineering design problems.
Lucian Buşoniu, Robert Babuška, Bart De Schutter
A comprehensive survey of multiagent reinforcement learning, covering formal goals, algorithms, and applications across cooperative, competitive, and mixed settings.
Unknown
Reinforcement learning boosts LLM performance on competitive programming and software engineering, comparing OpenAI models o1, o1-ioi, and o3.
Unknown
ConvNeXt modernizes standard ResNet with Transformer-inspired training and design, achieving competitive accuracy and scalability.
Unknown
Llama 3.1 adds multimodal capabilities via compositional cross-attention adapters for vision and speech, achieving competitive performance with GPT-4V and strong video reasoning.
Unknown
Gemma 2 improves open language models by interleaving local-global attention and using knowledge distillation, achieving competitive performance with much larger models.
Unknown
Meta's LLaMA introduces a collection of efficient foundation language models trained exclusively on publicly available data, achieving competitive performance with much larger models.
Shyamal Y. Dharia, Stephen D. Smith, Camilo E. Valderrama
Diff-Logic networks enable low-latency EEG classification on edge devices by compiling models into pure Boolean circuits, achieving competitive accuracy with constant inference time across scales.
Torres H., Juan P.
Proposes a Competitive Mixture of Experts (CMoE) framework using Functional Language Logic to replace probabilistic token prediction with deterministic mathematical reasoning for paradox-free AI.