Hacker News ·
SpacebarBench: Benchmarking Agents on Whiteboards
Submitted by zholer. Hacker News recorded 1 points and 0 comments as of Oct 21, 2026 (UTC).
Home / Browse
Models & research
Examine models, papers and evaluations with their methods and limitations.
802 public stories
Hacker News ·
Benchmarking Bot Detection Systems Against Modern AI Agents in 2026
Submitted by mdahardy. Hacker News recorded 44 points and 7 comments as of Oct 21, 2026 (UTC).
Hugging Face ·
FairStress: A deeper evaluation of fairness in LLM decision-making
Public article from Hugging Face. Open the source to read the full record.
Hacker News ·
Randomized test of an AI sales agent on 100 sites (ArXiv)
Hacker News recorded 1 points and 0 comments as of Oct 20, 2026 (UTC).
Hacker News ·
Emerging Trends in AI Agent Evaluation [pdf]
Hacker News recorded 1 points and 0 comments as of Oct 20, 2026 (UTC).
arXiv ·
Modeling Human Behavior in the Prisoner's Dilemma with Large Language Models
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Composable Parametric Demographic Injection for Values Survey Simulation
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Translating in style: An open-weight alternative to GPT-4 for culturally aligned machine translation
Public paper from arXiv. Open the source to read the full record.
arXiv ·
The Window Is in the Reasoning: When LLM Agents Decide to Follow Prompt Injections
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Discuss More, Comment Less: Boosting Automated Research via Multi-Perspective Collaborative Review
Public paper from arXiv. Open the source to read the full record.
arXiv ·
OR-License: Evaluating LLM Agents as Real-World Operations Research Practitioners
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Do LMs Know What Variables Do? Interpreting Code Representations via Variable Roles
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Prefix-Tuning for Arbitrary Output Sequences on Pretrained Transformers
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Preservation and loss of algorithmic structure in experimental diffusion of joint-action norms
Public paper from arXiv. Open the source to read the full record.
arXiv ·
CoDA: A Human-LM Collaborative Codebook Development Assistant for Domain-Specific Text Annotation
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Pronouncing Hard Words: Phoneme Adaptation of a Pretrained Conversational Text to Speech Model
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Prosodic Structure as a Critical Component in Phonemic Language Modelling at BabyLM Scale
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Multi-author style detection with stylistic embeddings
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Reasoning with Context Management and External Memory
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Language Models are Lazy Poets: Creative Mode Collapse at Scale
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Analyzing Repetition and Interference Effects in Memory-Constrained User Simulators
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Preserving Reasoning Diversity in Reinforcement Learning with Verifiable Rewards via Adaptive Replay
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Encoder-Free Multimodal LLMs Learn Implicit Encoders
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Can This LLM Annotate My Data? Estimating Zero-Shot LLM Performance
Public paper from arXiv. Open the source to read the full record.
arXiv ·
SLA-LENS: From Classroom Transcripts to CSLA-Informed and Evidence-Linked L2 Interaction Analytics
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Best of Both Worlds: Fine-Tuned Small Models Complement LLMs for Cost and Performance Benefits
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Modeling bilingual naming performance through laterally-connected SOMs
Public paper from arXiv. Open the source to read the full record.
arXiv ·
BanglaShield: A Script-Robust Safety Classifier and Benchmark for Bangla Native, Romanized, and Code-Mixed Prompts
Public paper from arXiv. Open the source to read the full record.
arXiv ·
Beyond Clean Accuracy: The Practical Challenges and Benefits of Pixel-Based Language Models
Public paper from arXiv. Open the source to read the full record.
arXiv ·
What Matters for NVFP4 Training? A Comparative Study of Low-Precision Pretraining Recipes
Public paper from arXiv. Open the source to read the full record.