NLI autonomously discovers a vocabulary of primitive operations and interprets variable-length programs via a neural executor, allowing end-to-end training and gradient-based test-time adaptation that outperforms prior methods on combinatorial generalization tasks.
hub Canonical reference
On the Measure of Intelligence
Canonical reference. 81% of citing Pith papers cite this work as background.
62
Pith papers citing it
Background
81% of classified citations
abstract
To make deliberate progress towards more intelligent and more human-like artificial systems, we need to be following an appropriate feedback signal: we need to be able to define and evaluate intelligence in a way that enables comparisons between two systems, as well as comparisons with humans. Over the past hundred years, there has been an abundance of attempts to define and measure intelligence, across both the fields of psychology and AI. We summarize and critically assess these definitions and evaluation approaches, while making apparent the two historical conceptions of intelligence that have implicitly guided them. We note that in practice, the contemporary AI community still gravitates towards benchmarking intelligence by comparing the skill exhibited by AIs and humans at specific tasks such as board games and video games. We argue that solely measuring skill at any given task falls short of measuring intelligence, because skill is heavily modulated by prior knowledge and experience: unlimited priors or unlimited training data allow experimenters to "buy" arbitrary levels of skills for a system, in a way that masks the system's own generalization power. We then articulate a new formal definition of intelligence based on Algorithmic Information Theory, describing intelligence as skill-acquisition efficiency and highlighting the concepts of scope, generalization difficulty, priors, and experience. Using this definition, we propose a set of guidelines for what a general AI benchmark should look like. Finally, we present a benchmark closely following these guidelines, the Abstraction and Reasoning Corpus (ARC), built upon an explicit set of priors designed to be as close as possible to innate human priors. We argue that ARC can be used to measure a human-like form of general fluid intelligence and that it enables fair general intelligence comparisons between AI systems and humans.
hub tools
citation-role summary
background 14
dataset 2
citation-polarity summary
claims ledger
- abstract To make deliberate progress towards more intelligent and more human-like artificial systems, we need to be following an appropriate feedback signal: we need to be able to define and evaluate intelligence in a way that enables comparisons between two systems, as well as comparisons with humans. Over the past hundred years, there has been an abundance of attempts to define and measure intelligence, across both the fields of psychology and AI. We summarize and critically assess these definitions and evaluation approaches, while making apparent the two historical conceptions of intelligence that h
- background depth transformers with this capability. These works have a similar aim to ours, enabling reasoning in latent space, but approach this goal from separate directions. For additional discussions related to the idea of construct- ing a prior that incentivizes reasoning and algorithm learn- ing at the expense of memorization of simple patterns, we also refer to Chollet (2019), Schwarzschild (2023), Li et al. (2020b) and Moulton (2023). 9. Future Work Aside from work extending and analyzing the scali
- background These techniques can be categorized into two main types based on the source of feedback: process reward models (PRMs) and prompted LLMs. The performance comparison are mainly shown in Table 4. Process Feedback from Process Rewarded Model Recent studies highlight the significance of feedback in developing effective PRMs for complex reasoning tasks, particularly in a step-level view [134, 423, 528]. (1) Process Annotated PRM Training: Earlier, Lightman et al. [449] demon- strate that training proc
co-cited works
representative citing papers
Flat minima are illusory; generalization is driven by weakness, a reparameterization-invariant measure of compatible completions that predicts performance better than sharpness on MNIST and Fashion-MNIST.
VisAnalog is a new controlled benchmark showing VLMs substantially underperform humans on visual concept transfer under one- to four-step deterministic transformations, with relation inference as the main failure mode.
EvoLib enables LLMs to accumulate, reuse, and evolve knowledge abstractions from inference trajectories at test time, yielding substantial gains on math reasoning, code generation, and agentic benchmarks without parameter updates or supervision.
The Divergent Remote Association Test (DRAT) is the first creativity test that significantly predicts LLMs' scientific ideation ability, unlike prior tests such as DAT or RAT.
Humans exhibit abstraction learning consistent with prospective compression of future tasks in non-stationary domains, unlike retrospective compression algorithms or LLM-based approaches.
An 800K-parameter Lattice Deduction Transformer reaches 100% accuracy on Sudoku-Extreme and Snowflake Sudoku and 99.9% on Maze-Hard by using lattice projections and abstract-interpretation supervision, while frontier LLMs score 0%.
Intervention complexity provides a family of canonical rewards indexed by resource bias that completes the Legg-Hutter framework and enables a two-dimensional view of intelligence as competence plus learning efficiency.
LLM agents execute scientific tasks but fail to follow core scientific reasoning norms such as evidence consideration and belief revision based on refutations.
CoT-PoT ensembling achieves self-consistency accuracy in LLMs with only two samples for 78.6% of tasks, reducing computation by 9.3x compared to standard methods.
A domain-independent analogy engine transfers Lean tactic patterns from probability to representation theory, producing four new machine-verified proofs.
The paper delivers the first survey of abductive reasoning in LLMs, a unified two-stage taxonomy, a compact benchmark, and an analysis of gaps relative to deductive and inductive reasoning.
ProofGrid is a new benchmark for LLM reasoning that uses machine-checkable proofs in minimal formal notation, revealing progress on basic tasks but major gaps in complex combinatorial and synthesis reasoning.
Factorization Regret measures how latent variable interactions affect performance, and RCCs enable learning them to achieve compositional generalization in partially observable tasks.
DecompSR is a large, symbolically verified benchmark dataset and generation framework that independently varies productivity, substitutivity, overgeneralisation, and systematicity to probe compositional multihop spatial reasoning in LLMs.
TRM with 7M parameters achieves 45% accuracy on ARC-AGI-1 and 8% on ARC-AGI-2, surpassing most LLMs with under 0.01% of their parameters.
VCBench is a new privacy-preserving benchmark showing LLMs like DeepSeek-V3 achieve over six times the market baseline precision in predicting founder success.
PuzzleWorld benchmark reveals state-of-the-art AI models solve only 18% of complex puzzlehunt problems with 40% stepwise accuracy, matching novices but trailing enthusiasts, while fine-tuning on traces yields modest gains.
PRIMETIME generator reveals that LLM datetime parsing and arithmetic primitives are individually unreliable but fully learnable via fine-tuning, enabling frontier-level accuracy on event planning with small LoRA models.
A recurrent-depth architecture enables language models to improve reasoning performance by iterating computation in latent space, achieving gains equivalent to much larger models on benchmarks.
LLMs display high variance and major accuracy drops on GSM-Symbolic variants of grade-school math problems, indicating they replicate training patterns rather than execute logical reasoning.
Meta Agent Search uses a meta-agent to iteratively program novel agentic systems in code, producing agents that outperform state-of-the-art hand-designed ones across coding, science, and math while transferring across domains and models.
Open-world evaluations using qualitative review of real-world tasks can give earlier warnings of frontier AI capabilities than automated benchmarks, as demonstrated by an AI agent publishing a simple iOS app with one minor human fix.
A universal LLM optimizer for text artifacts achieves SOTA results on six tasks including tripling ARC-AGI accuracy and cutting cloud costs by 40% via cross-task transfer and side information.
citing papers explorer
-
Gradient-Based Program Synthesis with Neurally Interpreted Languages
NLI autonomously discovers a vocabulary of primitive operations and interprets variable-length programs via a neural executor, allowing end-to-end training and gradient-based test-time adaptation that outperforms prior methods on combinatorial generalization tasks.
-
Are Flat Minima an Illusion?
Flat minima are illusory; generalization is driven by weakness, a reparameterization-invariant measure of compatible completions that predicts performance better than sharpness on MNIST and Fashion-MNIST.
-
VisAnalog: A Diagnostic Suite for Visual Concept Transfer on Natural Images
VisAnalog is a new controlled benchmark showing VLMs substantially underperform humans on visual concept transfer under one- to four-step deterministic transformations, with relation inference as the main failure mode.
-
Test-Time Learning with an Evolving Library
EvoLib enables LLMs to accumulate, reuse, and evolve knowledge abstractions from inference trajectories at test time, yielding substantial gains on math reasoning, code generation, and agentic benchmarks without parameter updates or supervision.
-
Assessing the Creativity of Large Language Models: Testing, Limits, and New Frontiers
The Divergent Remote Association Test (DRAT) is the first creativity test that significantly predicts LLMs' scientific ideation ability, unlike prior tests such as DAT or RAT.
-
Prospective Compression in Human Abstraction Learning
Humans exhibit abstraction learning consistent with prospective compression of future tasks in non-stationary domains, unlike retrospective compression algorithms or LLM-based approaches.
-
Lattice Deduction Transformers
An 800K-parameter Lattice Deduction Transformer reaches 100% accuracy on Sudoku-Extreme and Snowflake Sudoku and 99.9% on Maze-Hard by using lattice projections and abstract-interpretation supervision, while frontier LLMs score 0%.
-
Intervention Complexity as a Canonical Reward and a Measure of Intelligence
Intervention complexity provides a family of canonical rewards indexed by resource bias that completes the Legg-Hutter framework and enables a two-dimensional view of intelligence as competence plus learning efficiency.
-
AI scientists produce results without reasoning scientifically
LLM agents execute scientific tasks but fail to follow core scientific reasoning norms such as evidence consideration and belief revision based on refutations.
-
Self-Consistency from Only Two Samples: CoT-PoT Ensembling for Efficient LLM Reasoning
CoT-PoT ensembling achieves self-consistency accuracy in LLMs with only two samples for 78.6% of tasks, reducing computation by 9.3x compared to standard methods.
-
Yanasse: Finding New Proofs from Deep Vision's Analogies, Part 1
A domain-independent analogy engine transfers Lean tactic patterns from probability to representation theory, producing four new machine-verified proofs.
-
Wiring the 'Why': A Unified Taxonomy and Survey of Abductive Reasoning in LLMs
The paper delivers the first survey of abductive reasoning in LLMs, a unified two-stage taxonomy, a compact benchmark, and an analysis of gaps relative to deductive and inductive reasoning.
-
Stress-Testing the Reasoning Competence of LLMs With Proofs Under Minimal Formalism
ProofGrid is a new benchmark for LLM reasoning that uses machine-checkable proofs in minimal formal notation, revealing progress on basic tasks but major gaps in complex combinatorial and synthesis reasoning.
-
Factorization Regret mediates compositional generalization in latent space
Factorization Regret measures how latent variable interactions affect performance, and RCCs enable learning them to achieve compositional generalization in partially observable tasks.
-
DecompSR: A dataset for decomposed analyses of compositional multihop spatial reasoning
DecompSR is a large, symbolically verified benchmark dataset and generation framework that independently varies productivity, substitutivity, overgeneralisation, and systematicity to probe compositional multihop spatial reasoning in LLMs.
-
Less is More: Recursive Reasoning with Tiny Networks
TRM with 7M parameters achieves 45% accuracy on ARC-AGI-1 and 8% on ARC-AGI-2, surpassing most LLMs with under 0.01% of their parameters.
-
VCBench: Benchmarking LLMs in Venture Capital
VCBench is a new privacy-preserving benchmark showing LLMs like DeepSeek-V3 achieve over six times the market baseline precision in predicting founder success.
-
PuzzleWorld: A Benchmark for Multimodal, Open-Ended Reasoning in Puzzlehunts
PuzzleWorld benchmark reveals state-of-the-art AI models solve only 18% of complex puzzlehunt problems with 40% stepwise accuracy, matching novices but trailing enthusiasts, while fine-tuning on traces yields modest gains.
-
PRIMETIME : Limits of LLMs in Temporal Primitives
PRIMETIME generator reveals that LLM datetime parsing and arithmetic primitives are individually unreliable but fully learnable via fine-tuning, enabling frontier-level accuracy on event planning with small LoRA models.
-
Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
A recurrent-depth architecture enables language models to improve reasoning performance by iterating computation in latent space, achieving gains equivalent to much larger models on benchmarks.
-
GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models
LLMs display high variance and major accuracy drops on GSM-Symbolic variants of grade-school math problems, indicating they replicate training patterns rather than execute logical reasoning.
-
Automated Design of Agentic Systems
Meta Agent Search uses a meta-agent to iteratively program novel agentic systems in code, producing agents that outperform state-of-the-art hand-designed ones across coding, science, and math while transferring across domains and models.
-
Open-World Evaluations for Measuring Frontier AI Capabilities
Open-world evaluations using qualitative review of real-world tasks can give earlier warnings of frontier AI capabilities than automated benchmarks, as demonstrated by an AI agent publishing a simple iOS app with one minor human fix.
-
optimize_anything: A Universal API for Optimizing any Text Parameter
A universal LLM optimizer for text artifacts achieves SOTA results on six tasks including tripling ARC-AGI accuracy and cutting cloud costs by 40% via cross-task transfer and side information.
-
Generative Recursive Reasoning
GRAM is a latent-variable generative model that performs recursive reasoning via stochastic trajectories, trained with amortized variational inference to support multi-hypothesis reasoning and unconditional generation.
-
LEAP: Trajectory-Level Evaluation of LLMs in Iterative Scientific Design
LEAPBench shows trajectory scoring changes best-model rankings on 53% of tasks, LLMs do not beat Bayesian optimization, and domain-aware prompting underperforms domain-agnostic on biology tasks aligned with published literature.
-
The Evaluation Trap: Benchmark Design as Theoretical Commitment
AI benchmarks trap progress by operationalizing assumptions that redefine capabilities around the benchmarks themselves, and Epistematics provides an audit procedure to detect when evaluations cannot discriminate claimed capabilities from proxy behaviors.
-
The Generalized Turing Test: A Foundation for Comparing Intelligence
The Generalized Turing Test defines relative intelligence as the inability of one agent to distinguish an imitator from the original through interaction.
-
When to Re-Commit: Temporal Abstraction Discovery for Long-Horizon Vision-Language Reasoning
Learns state-conditioned commitment depth in a 7B vision-language policy that jointly predicts actions and replan intervals, outperforming fixed-depth baselines and larger models on Sliding Puzzle and Sokoban while providing a theoretical dominance result.
-
Forge: Quality-Aware Reinforcement Learning for NP-Hard Optimization in LLMs
OPT-BENCH trains LLMs on NP-hard optimization via quality-aware RLVR, achieving 93.1% success rate and 46.6% quality ratio on Qwen2.5-7B while outperforming GPT-4o and transferring gains to other domains.
-
Continuous Latent Diffusion Language Model
Cola DLM proposes a hierarchical latent diffusion model that learns a text-to-latent mapping, fits a global semantic prior in continuous space with a block-causal DiT, and performs conditional decoding, establishing latent prior modeling as an alternative to token-level autoregressive language model
-
One Step Forward and K Steps Back: Better Reasoning with Denoising Recursion Models
Denoising Recursion Models train multi-step noise reversal in looped transformers and outperform the prior Tiny Recursion Model on ARC-AGI.
-
Representation-Guided Parameter-Efficient LLM Unlearning
REGLU guides LoRA-based unlearning via representation subspaces and orthogonal regularization to outperform prior methods on forget-retain trade-off in LLM benchmarks.
-
C-voting: Confidence-Based Test-Time Voting without Explicit Energy Functions
C-voting improves recurrent reasoning models by selecting among multiple latent trajectories the one with highest average top-1 probability, achieving 4.9% better Sudoku-hard accuracy than energy-based voting and outperforming HRM on Sudoku-extreme and Maze when paired with the new ItrSA++ model.
-
ARC-AGI-3: A New Challenge for Frontier Agentic Intelligence
ARC-AGI-3 is a benchmark where humans solve 100% of tasks but frontier AI systems score below 1% as of March 2026, using efficiency-based scoring grounded in human baselines.
-
ScaLoRA: Optimally Scaled Low-Rank Adaptation for Efficient High-Rank Fine-Tuning
ScaLoRA analytically derives per-update column scalings that let low-rank increments accumulate into high-rank weight updates, yielding faster convergence and higher accuracy than prior LoRA variants on LLMs up to 12B parameters.
-
Test-Time Matching: Unlocking Compositional Reasoning in Multimodal Models
Introduces group matching score for better evaluation of compositional reasoning and Test-Time Matching (TTM) algorithm for unsupervised self-improvement in multimodal models, achieving SOTA gains including surpassing GPT-4.1 and estimated human performance.
-
AInstein: Can LLMs Solve Research Problems From Parametric Memory Alone?
LLMs generate valid solutions to over 70% of AI research problems from parametric memory alone but rediscover the exact published approach less than 19% of the time, with performance limited by cross-domain analogical transfer.
-
Video models are zero-shot learners and reasoners
Generative video models exhibit emergent zero-shot capabilities across perception, manipulation, and basic reasoning tasks.
-
Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning
High-entropy minority tokens drive RLVR gains, so restricting gradients to the top 20% maintains or improves performance over full updates on Qwen3 models, especially larger ones.
-
Large Language Monkeys: Scaling Inference Compute with Repeated Sampling
Repeated sampling scales problem coverage log-linearly with sample count, improving SWE-bench Lite performance from 15.9% to 56% using 250 samples.
-
Probabilistic Tiny Recursive Model
PTRM adds stochastic Gaussian noise to Tiny Recursive Model recursion for parallel trajectory exploration and Q-head selection, raising Sudoku-Extreme accuracy from 87.4% to 98.75% and Pencil Puzzle Bench from 62.6% to 91.2% without retraining.
-
Predicting Performance of Symbolic and Prompt Programs with Examples
Proposes RAP, a retrieval-based approximate prior method, to predict performance of symbolic programs and LLM prompts on new tasks using a Bernoulli model and corpus-derived performance distributions.
-
Deep Vision: A Formal Proof of Wolstenholmes Theorem in Lean 4
Wolstenholme's theorem is formally verified in Lean 4 via expansion of a shifted factorial product and vanishing power sums modulo p.
-
The Rise and Fall of $G$ in AGI
PCA on AI model benchmarks reveals a general intelligence factor that rises then falls as specialized reasoning models appear, inverting the expected move toward parsimonious mechanisms.
-
Kuramoto Oscillatory Phase Encoding: Neuro-inspired Synchronization for Improved Learning Efficiency
KoPE adds Kuramoto-based oscillatory phase states and synchronization to Vision Transformers, improving training, parameter, and data efficiency on structured vision tasks.
-
From Pixels to Digital Agents: An Empirical Study on the Taxonomy and Technological Trends of Reinforcement Learning Environments
An empirical literature analysis reveals a bifurcation in RL environments into Semantic Prior (LLM-dominated) and Domain-Specific Generalization ecosystems with distinct cognitive fingerprints.
-
Intelligence Inertia: Physical Isomorphism and Applications
Intelligence Inertia models the computational resistance to structural change in neural networks via a heuristic relativistic analogy, yielding a J-shaped cost curve that diverges from classical approximations.
-
How Psychological Learning Paradigms Shaped and Constrained Artificial Intelligence
AI's compositional reasoning failures originate in psychological learning paradigms that shaped its architectures, and the ReSynth trimodular framework is proposed to embed systematicity structurally.
-
Position: AI Evaluations Should be Grounded on a Theory of Capability
AI evaluations should be reframed as inference tasks grounded in an explicit theory of capability, with an empirical demonstration that results depend on modeling assumptions and a proposed Evaluation Card for transparency.