CogGym is a scalable, unified framework that standardizes diverse cognitive experiments into a task‑agnostic Experiment Markup Language (EML) for systematic comparison of human and AI behavior. The initial release curates 258 experiments from 100 papers focused on human commonsense reasoning and evaluates 50 large language models, revealing a scaling trend where larger models better reproduce human judgments but still lag far behind human split‑half reliability. The framework aims to continually incorporate new cognitive science experiments to track where model behavior aligns with or diverges from human cognition as models evolve.
arXiv:2502. 20502v2 Announce Type: replace Abstract: Recent advances in Artificial Intelligence (AI) have yielded powerful computational models that, by learning from vast amounts of human-generated data, are increasingly posited as approximate models of human cognition.
arXiv:2605.06524v3 Announce Type: replace Abstract: Reliable human-machine discrimination is becoming increasingly important as Large Language Models and autonomous agents are deployed in online sett...
arXiv:2509. 14704v3 Announce Type: replace Abstract: Benchmark saturation and training-data contamination increasingly obscure whether reported gains in large language models (LLMs) reflect genuine advances in reasoning or familiarity with recurring patterns in benchmark problems.
The paper introduces WROP, a dataset of 150 hand‑designed cognitive tasks aimed at testing object permanence in video generation models. Using Blender, the authors generate over 10,000 samples per task, compiling a 1.5 M‑sample training corpus and a 300‑question exam. They evaluate 14 video models, showing that their 16B world model, PWM‑WROP, ranks first among continuation models and third overall in a blind Elo study.
arXiv:2602.00685v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly used as simulated participants in social science experiments, but their behavior is often unstable an...
WordPolo is a word‑finding task that evaluates language models by having them guess an unknown target word and receive semantic similarity feedback. Participants start with no knowledge, make iterative guesses, and receive distance scores that guide them through semantic space. The study tests recent LLMs, LRMs, humans, and a heuristic on 1,500 puzzles, revealing that while solve rates vary widely, many models make meaningful progress and exhibit human‑like strategies, highlighting the importance of assessing reasoning processes, not just final accuracy.
arXiv:2601. 02854v2 Announce Type: replace Abstract: As an agent-level reasoning and coordination paradigm, Multi-Agent Debate (MAD) orchestrates multiple agents through structured debate to improve answer quality and support complex reasoning.
arXiv:2602. 12279v2 Announce Type: replace-cross Abstract: Unified models can handle both multimodal understanding and generation within a single architecture, yet they typically operate in a single pass without iteratively refining their outputs.
LiveMathematicianBench is a dynamic multiple‑choice benchmark for research‑level mathematical reasoning, built from recent arXiv papers published after model training cutoffs. It introduces a thirteen‑category logical taxonomy of theorem types and uses a proof‑sketch‑guided distractor pipeline to create plausible but invalid answer choices, enhancing sensitivity to genuine reasoning. Evaluation shows current large language models perform poorly, with the best model scoring 43.5% overall and only 17.6% under substitution‑resistant conditions, indicating the benchmark’s difficulty and realism.
CoVA‑SFT is a new large‑scale dataset comprising 51.9K samples and over 222K multimodal reasoning steps that teach models to interleave text and visual abstractions across five layout families and 17 complex tasks. It includes explicit rationale formulations, agentic renderings, and verification loops to help models build and maintain internal visual workspaces for purely textual reasoning problems. A companion benchmark, CoVA‑Bench, contains 1,700 held‑out test samples for reproducible evaluation, and models fine‑tuned on CoVA‑SFT outperform all interleaved CoT baselines by more than 2× on average, though they still lag behind strong text‑only CoT baselines.
The paper introduces the concept of an agent’s "taste"—its ability to make effective long‑horizon decisions—and presents Taste‑Bench, a new benchmark that automatically generates decision‑fork questions from agent trajectories. Taste‑Bench evaluates models on choosing the best path without seeing future outcomes, revealing that top models answer only about 60% of questions correctly and that later‑appearing evidence makes forks harder. The authors also demonstrate that training a student model to mimic a teacher’s judgment improves decision quality and overall success on held‑out software engineering tasks.
arXiv:2607. 08317v1 Announce Type: new Abstract: Modern AI models achieve strong performance on many established benchmarks, yet they still fail on tasks that humans find almost trivial, such as manipulating a string or drawing a dog with five legs.