| 2641. |
Query Agnostic Adversarial Triggers for Reasoning Models (arxiv.org) |
|
3 points by fzliu on Jul 29, 2025 | hide | past | pdf | discuss
|
|
| 2642. |
Language Model Can Be a Steganographic Privacy Leaking Agent (arxiv.org) |
|
3 points by dennis-tra on Jul 29, 2025 | hide | past | pdf | discuss
|
|
| 2643. |
Supervised fine tuning on curated data is reinforcement learning (arxiv.org) |
|
71 points by GabrielBianconi on Jul 29, 2025 | hide | past | pdf | 19 comments
|
|
| 2644. |
TrimLLM: Progressive Layer Dropping for Domain-Specific LLMs (arxiv.org) |
|
2 points by pulkitsh1234 on Jul 29, 2025 | hide | past | pdf | discuss
|
|
| 2645. |
SmallThinker: A Family of Efficient LLMs Natively Trained for Local Deployment (arxiv.org) |
|
2 points by limoce on Jul 29, 2025 | hide | past | pdf | discuss
|
|
| 2646. |
Learning without training: The implicit dynamics of in-context learning (arxiv.org) |
|
2 points by JnBrymn on Jul 28, 2025 | hide | past | pdf | discuss
|
|
| 2647. |
A Unified Frontier in Neuroscience, AI and Neuromorphic Systems (arxiv.org) |
|
3 points by belter on Jul 28, 2025 | hide | past | pdf | discuss
|
|
| 2648. |
Plex: Perturbation-Free Local Explanations for LLM-Based Text Classification (arxiv.org) |
|
2 points by PaulHoule on Jul 28, 2025 | hide | past | pdf | discuss
|
|
| 2649. |
Self-attention transforms a prompt into a low-rank weight-update (arxiv.org) |
|
17 points by Labo333 on Jul 28, 2025 | hide | past | pdf | 1 comment
|
|
| 2650. |
GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning (arxiv.org) |
|
8 points by LakshyAAAgrawal on Jul 28, 2025 | hide | past | pdf | discuss
|
|
| 2651. |
Dynamic Chunking for End-to-End Hierarchical Sequence Modeling (arxiv.org) |
|
3 points by mromanuk on Jul 27, 2025 | hide | past | pdf | discuss
|
|
| 2652. |
Every Model Learned by Gradient Descent Is Approximately a Kernel Machine (arxiv.org) |
|
4 points by LordNibbler on Jul 27, 2025 | hide | past | pdf | discuss
|
|
| 2653. |
Why Neural Networks Can Discover Symbolic Structures (arxiv.org) |
|
9 points by calebkaiser on Jul 27, 2025 | hide | past | pdf | discuss
|
|
| 2654. |
Does visualization help AI understand data? (arxiv.org) |
|
3 points by babushkaboi on Jul 27, 2025 | hide | past | pdf | discuss
|
|
| 2655. |
Hierarchical Reasoning Model (arxiv.org) |
|
339 points by hansmayer on Jul 27, 2025 | hide | past | pdf | 106 comments
|
|
| 2656. |
AlphaGo Moment for Model Architecture Discovery (arxiv.org) |
|
38 points by Jimmc414 on Jul 26, 2025 | hide | past | pdf | 7 comments
|
|
| 2657. |
Market-Derived Financial Sentiment Analysis: Context-Aware Language Models (arxiv.org) |
|
5 points by Bluestein on Jul 26, 2025 | hide | past | pdf | discuss
|
|
| 2658. |
The Sparse Frontier: Sparse Attention Trade-Offs in Transformer LLMs (arxiv.org) |
|
6 points by Bogdanp on Jul 26, 2025 | hide | past | pdf | discuss
|
|
| 2659. |
A Fact-Grounded Multimodal Writing Assistant Based on Offline Knowledge Base (arxiv.org) |
|
2 points by PaulHoule on Jul 25, 2025 | hide | past | pdf | discuss
|
|
| 2660. |
Learning without training: The implicit dynamics of in-context learning (arxiv.org) |
|
1 point by simonpure on Jul 25, 2025 | hide | past | pdf | discuss
|
|
| 2661. |
Explainable Mapper: Charting LLM Embedding Spaces Using Perturbation-Based (arxiv.org) |
|
2 points by badmonster on Jul 25, 2025 | hide | past | pdf | 1 comment
|
|
| 2662. |
TaxCalcBench: Evaluating Frontier Models on the Tax Calculation Task (arxiv.org) |
|
2 points by sundaypancakes on Jul 25, 2025 | hide | past | pdf | discuss
|
|
| 2663. |
WhoFi: Deep Person Re-Identification via Wi-Fi Channel Signal Encoding (arxiv.org) |
|
55 points by wut42 on Jul 25, 2025 | hide | past | pdf | 10 comments
|
|
| 2664. |
Setol: SemiEmpirical Theory of (Deep) Learning (arxiv.org) |
|
6 points by charleshmartin on Jul 25, 2025 | hide | past | pdf | 3 comments
|
|
| 2665. |
Gemini 2.5 Pro Capable of Winning Gold at IMO 2025 (arxiv.org) |
|
1 point by chaosprint on Jul 25, 2025 | hide | past | pdf | 1 comment
|
|
| 2666. |
A Mixture of Experts Approach to Handle Concept Drifts (arxiv.org) |
|
1 point by adbabdadb on Jul 25, 2025 | hide | past | pdf | 1 comment
|
|
| 2667. |
TaxCalcBench: Can AI file your taxes? (not yet) (arxiv.org) |
|
1 point by michaelrbock on Jul 25, 2025 | hide | past | pdf | discuss
|
|
| 2668. |
WhoFi: Deep Person Re-Identification via Wi-Fi Channel Signal Encoding (arxiv.org) |
|
4 points by jonbaer on Jul 24, 2025 | hide | past | pdf | discuss
|
|
| 2669. |
Safer AI Agents Through Understanding and Evaluating Mobile UI Operation Impacts (arxiv.org) |
|
2 points by azhenley on Jul 24, 2025 | hide | past | pdf | discuss
|
|
| 2670. |
The Implicit Dynamics of In-Context Learning (arxiv.org) |
|
1 point by omarsar on Jul 24, 2025 | hide | past | pdf | discuss
|
|
|
| More |
