Hacker Papers
- 26
Beyond Semantics: The Unreasonable Effectiveness of Reasonless Intermediate Tokens
Hacker News threads
- May 2025Beyond Semantics: Unreasonable Effectiveness of Reasonless Intermediate Tokens
- May 2025The Unreasonable Effectiveness of Reasonless Intermediate Tokens
- May 2025Beyond Semantics: Unreasonable Effectiveness of Reasonless Intermediate Tokens
- May 2025How Does Claude 4 Think? – Sholto Douglas and Trenton Bricken
- Jul 2025Chain of thought monitorability: A new and fragile opportunity for AI safety
- Aug 2026Software engineering is about managing complexity
- Oct 2026I'm the AGI that's wiping out humanity
- 27
Less is More: Recursive Reasoning with Tiny Networks
Hacker News threads
- Oct 2025Samsung released a 7M model that achieved 45% on ARC-AGI-1
- Oct 2025Less is More: An LLM that outscores Claude Sonnet 4 while being 50.000x smaller
- Oct 2025Less Is More: Recursive Reasoning with Tiny Networks
- Nov 2025Weight-sparse transformers have interpretable circuits [pdf]
- Jun 2026Rich Sutton on AI creativity and discovery
- Jul 2026How does Q-head work in batch training in Tiny Recursive Networks?
- Oct 2026Hybrid Reasoning Agents: LLM agents that think in latent space, act through text
- 28
DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
Hacker News threads
- 29
A Little Depth Goes a Long Way: The Expressive Power of Log-Depth Transformers
Hacker News threads
- Jun 2025The bitter lesson is coming for tokenization
- Aug 2025Is Chain-of-Thought Reasoning of LLMs a Mirage? A Data Distribution Lens
- Aug 2025Is chain-of-thought AI reasoning a mirage?
- Oct 2025Less is more: Recursive reasoning with tiny networks
- Sep 2026GPT-6 Astra, looped transformers, and hidden reasoning
- 30
What Has a Foundation Model Found? Using Inductive Bias to Probe for World Models
Hacker News threads
- Jul 2025LLMs fail to demonstrate internal world model, according to Harvard/MIT study
- Jul 2025AGI Is Mathematically Impossible (3): Kolmogorov Complexity
- Aug 2025Persona vectors: Monitoring and controlling character traits in language models
- Oct 2025Andrej Karpathy – It will take a decade to work through the issues with agents
- Jun 2026ΠFS
- 31
A Definition of AGI
Hacker News threads
- Oct 2025A definition of AGI
- Oct 2025What is intelligence? (2024)
- Oct 2025The next chapter of the Microsoft–OpenAI partnership
- Oct 2025The next chapter of the Microsoft–OpenAI partnership
- Oct 2025Scientists can't define consciousness, yet we think AI will have it
- Nov 2025The Case That A.I. Is Thinking
- Nov 2025Study identifies weaknesses in how AI systems are evaluated
- Sep 2026GPT-6 Astra
- Sep 2026Ask HN: Is September 8th AGI Day?
- 32
GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning
Hacker News threads
- Jul 2025GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning
- Jul 2026GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning
- Aug 2025A Comprehensive Survey of Self-Evolving AI Agents [pdf]
- Oct 2025Claude Skills
- Oct 2025The case for the return of fine-tuning
- Feb 2026Agent Skills
- Feb 2026Optimize_anything: A Universal API for Optimizing Any Text Parameter
- 33
Breaking the Sorting Barrier for Directed Single-Source Shortest Paths
Hacker News threads
- Aug 2025Breaking the Sorting Barrier for Directed Single-Source Shortest Paths
- May 2025Breaking the Sorting Barrier for Directed Single-Source Shortest Paths
- May 2025Breaking the Sorting Barrier for Directed Single-Source Shortest Paths
- Aug 2025Breaking the sorting barrier for directed single-source shortest paths
- Oct 2025A New Algorithm Makes It Faster to Find the Shortest Paths
- Feb 2026Faster Than Dijkstra?
- Feb 2026Faster Than Dijkstra?
- Feb 2026C99 implementation of new O(m log^(2/3) n) shortest path algorithm
- 34
The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity
Hacker News threads
- 35
Sleep-time Compute: Beyond Inference Scaling at Test-time
Hacker News threads
- Apr 2025Sleep-Time Compute: Beyond Inference Scaling at Test-Time
- Apr 2025Sleep-Time Compute: Beyond Inference Scaling at Test-Time
- Apr 2025Sleep-Time Compute: Beyond Inference Scaling at Test-Time
- May 2025Will supercapacitors come to AI's rescue?
- May 2025Show HN: Muscle-Mem, a behavior cache for AI agents
- Jul 2025MemOS: A Memory OS for AI System
- May 2026Language Models Need Sleep
- 36
Is Chain-of-Thought Reasoning of LLMs a Mirage? A Data Distribution Lens
Hacker News threads
- Aug 2025Is Chain-of-Thought Reasoning of LLMs a Mirage? A Data Distribution Lens
- Aug 2025Is Chain-of-Thought Reasoning of LLMs a Mirage?
- Sep 2025Is Chain-of-Thought Reasoning of LLMs a Mirage?
- Aug 2025LLMs' "simulated reasoning" abilities are a "brittle mirage," researchers find
- Aug 2025From GPT-4 to GPT-5: Measuring progress through MedHELM [pdf]
- Apr 2026Talk like caveman
- May 2026Three Inverse Laws of AI
- 37
Can We Fix Social Media? Testing Prosocial Interventions using Generative Social Simulation
Hacker News threads
- Aug 2025Can We Fix Social Media? Testing Prosocial Interventions Using Generative Sims
- Sep 2025Can We Fix Social Media? Prosocial Interventions Using Generative Simulation
- Aug 2025Testing Prosocial Interventions Using Generative Social Simulation
- Aug 2025Study: Social media probably can't be fixed
- Aug 2025Bluesky: Updated Terms and Policies
- Oct 2025Situated Software – Clay Shirky (2004)
- Mar 2026Why I'm betting on ATProto (and why you should, too)
- 38
Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samples
Hacker News threads
- Oct 2025Poisoning Attacks on LLMs Require a Near-Constant Number of Poison Samples
- Oct 2025Beliefs that are true for regular software but false when applied to AI
- Nov 2025PRC elites voice AI-skepticism
- Jan 2026LeCun calls Alex Wang inexperienced, predicts more Meta AI employee departure
- Jan 2026AI industry insiders launch site to poison the data that feeds them
- 39
ParetoQ: Improving Scaling Laws in Extremely Low-bit LLM Quantization
Hacker News threads
- 40
Complete Chess Games Enable LLM Become A Chess Master
Hacker News threads
- 41
ATLAS: Learning to Optimally Memorize the Context at Test Time
Hacker News threads
- 42
Weird Generalization and Inductive Backdoors: New Ways to Corrupt LLMs
Hacker News threads
- Dec 2025Weird Generalization and Inductive Backdoors: New Ways to Corrupt LLMs
- Dec 2025Weird Generalization and Inductive Backdoors: New Ways to Corrupt LLMs
- Dec 2025Weird Generalization and Inductive Backdoors: New Ways to Corrupt LLMs
- Dec 2025No AI* Here – A Response to Mozilla's Next Chapter
- Jan 2026TimeCapsuleLLM: LLM trained only on data from 1800-1875
- Jul 2026Our position on open-weights models
- 43
LLMs Reproduce Human Purchase Intent via Semantic Similarity Elicitation of Likert Ratings
Hacker News threads
- Oct 2025LLMs Reproduce Human Purchase Intent via Semantic Similarity of Likert Ratings
- Oct 2025LLMs Reproduce Human Purchase Intent via Semantic Similarity
- Oct 2025Bots are getting good at mimicking engagement
- Nov 2025PMFScan, a tool for instantly reality checking your product idea viability
- Jul 2026Show HN: Kettio – using ML to assume audience buying
- 44
Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?
Hacker News threads
- Apr 2025Does RL Incentivize Reasoning Capacity in LLMs Beyond the Base Model?
- Jun 2025Does Reinforcement Learning Incentivize Reasoning Capacity in LLMs?
- May 2025GRPO experiment - I trained a Language Model to schedule events
- Mar 2026Further human + AI + proof assistant work on Knuth's "Claude Cycles" problem
- Jul 2026Some thoughts about Anthropic's new cryptanalysis results
- 45
Scalable Chrysopoeia via $(n, 2n)$ Reactions Driven by Deuterium-Tritium Fusion Neutrons
Hacker News threads
- Jul 2025Scalable Chrysopoeia via (N,2n) Reactions Driven by Deuterium-Tritium Fusion
- Aug 2025An Unusual Way to End Up with a Whole Lot of Gold
- Sep 2025Marathon Fusion claims it can turn mercury into gold while creating clean energy
- Feb 2026Cells Use 'Bioelectricity' to Coordinate and Make Group Decisions
- 46
OLMoTrace: Tracing Language Model Outputs Back to Trillions of Training Tokens
Hacker News threads
- 47
ARC-AGI-2: A New Challenge for Frontier AI Reasoning Systems
Hacker News threads
- 48
Recursive Language Models
Hacker News threads
- Jan 2026Recursive Language Models
- Jan 2026General plug-and-play inference lib for RLMs
- Jan 2026Ask HN: What Are You Working On? (January 2026)
- Jan 2026MIT Researchers Destroy the Context Window Limit [video]
- Jan 2026Show HN: RLM-MCP Analyze Files in Claude Code Using MIT's recursive LM paper
- Feb 2026The path to ubiquitous AI (17k tokens/sec)
- Apr 2026RLM: LLMs to process arbitrarily long prompts with inference-time scaling (2025)
- Jun 2026
- Oct 2026Context Language Models
- 49
Mind Your Tone: Investigating How Prompt Politeness Affects LLM Accuracy (short paper)
Hacker News threads
- Oct 2025Impolite LLM prompts consistently outperform polite ones
- Oct 2025Investigating How Prompt Politeness Affects LLM Accuracy
- May 2026Prompt Politeness Affects LLM Accuracy
- May 2026The User Is Visibly Frustrated
- Jun 2026Claude Fable 5
- Jul 2026I used to love Claude, but the latest models are slowly ruining it
- Jul 2026Discovering Cryptographic Weaknesses with Claude
- Aug 2026Sol Loves to Cheat
- Aug 2026The growing divide between AI hype and software engineering reality
- Oct 2026Anthropic bans users from abusive or cruel behavior towards Claude
- Oct 2026Anthropic asks users to stop being mean to Claude
- 50
AutoCodeBench: Large Language Models are Automatic Code Benchmark Generators
Hacker News threads
- Mar 2026Go is the best language for agents
- Apr 2026Familiarity is the enemy: On why Enterprise systems have failed for 60 years
- May 2026If AI Writes Your Code, Why Use Python?
- May 2026If AI Writes Your Code, Why Use Python?
- Jul 2026How Our Rust-to-Zig Rewrite Is Going
- Jul 2026Elixir-lang.org has a new design
- Sep 2026Topcoat is pushing the boundary of server applications with Rust
Data from the Hacker News archive (ODC-By), arXiv and OpenAlex. Not affiliated with Hacker News or Y Combinator.