Papers Categories Discussions Trending Leaderboard

Discussions

Grounded conversations about papers — questions, critiques, reproduction notes, and ideas.

All Question Critique Reproduction Explanation Research Idea

Tag: ML Systems clear

Tags

Accounting0 Aerospace Engineering0 Algebra0 Analysis0 Analytical Chemistry0 Anthropology0 Art & Design0 Artificial Intelligence0 Astrophysics0 Atmospheric Science0 Audio & Speech0 Behavioral Science0 Bioinformatics12 Biological Physics0 Biomedical Engineering0 Biomolecules0 Causal Inference0 Cell Behavior0 Cell Biology0 Chemical Engineering0 Civil Engineering0 Climate Science0 Clinical Medicine0 Clinical Psychology0 Cognitive Science4 Combinatorics0 Communication & Media0 Computational Chemistry0 Computational Complexity0 Computational Engineering0 Computer Architecture0 Computers & Society0 Computer Vision8 Condensed Matter0 Databases0 Data Structures & Algorithms0 Demography0 Developmental Psychology0 Digital Libraries0 Dynamical Systems0 Ecology0 Econometrics0 Economics6 Education0 EHR2 Emerging Technologies0 Energy0 Entrepreneurship0 Environmental Science0 Ethics0 Evolution0 Finance0 Fluid Dynamics0 Formal Languages0 General Physics0 General Relativity0 Genetics0 Genomic Medicine0 Genomics0 Geography0 Geology0 Geometry & Topology0 Geophysics0 Graphics0 HCI0 High Energy Physics0 History0 Image & Video0 Industrial Engineering0 Information Retrieval0 Information Theory0 Inorganic Chemistry0 Law0 Library & Information Science0 Linguistics0 Literature0 Logic0 Logic in CS0 Machine Learning0 Management0 Marketing0 Materials Chemistry0 Materials Science0 Mathematical Physics0 Mechanical Engineering0 Medical Imaging0 Medicine2 Mental Health0 Microbiology0 ML Systems14 Molecular Biology0 Molecular Networks0 Multiagent Systems0 Multimedia0 Nanotechnology0 Networking0 Neural & Evolutionary Computing0 Neuroscience3 NLP28 Nonlinear Sciences0 Nuclear Physics0 Number Theory0 Numerical Analysis0 Oceanography0 Operating Systems0 Operations Research0 Optics0 Optimization0 Optimization & Control0 Organic Chemistry0 Pharmacology0 Philosophy0 Physical Chemistry0 Political Science0 Polymers0 Populations & Evolution0 Portfolio Management0 Pricing of Securities0 Probability0 Programming Languages0 Psychology4 Public Health3 Public Health & Epidemiology0 Public Policy0 Quantitative Finance0 Quantum Physics0 Religion0 Risk Management0 Robotics0 Security0 Semiconductors0 Signal Processing0 Social & Information Networks0 Social Psychology0 Sociology3 Software Engineering0 Sound0 Statistics10 Structural Biology0 Symbolic Computation0 Systems & Control0 Theory11 Tissues & Organs0 Trading & Microstructure0

PaperTogether— Discuss papers together.

Papers Discussions Community Guidelines Privacy Terms FeedbackMetadata & discussion only — never copyrighted PDFs.

Reproductionon Denoising Diffusion Probabilistic Models

Reproduction note: training seems sensitive to seed

Computer Vision ML Systems

4 2about 1 month ago

by tbecker

Reproductionon Chain-of-Thought Prompting Elicits Reasoning in Large Language Models

Reproduction note: GSM8K with GPT-3.5-turbo

5 1about 1 month ago

by tbecker

Critiqueon Scaling Laws for Neural Language Models

The compute-optimal frontier ignores inference cost

3 2about 1 month ago

by tbecker

Critiqueon Attention Is All You Need

Hot take: 'scale is all you need' is becoming an ideology — are we hitting a wall?

ML Systems Theory

4 5about 1 month ago

by tbecker

Reproductionon Denoising Diffusion Probabilistic Models

DDPM 재현 중 시드별 FID 분산이 생각보다 큽니다

Computer Vision ML Systems

2 3about 2 months ago

by minseok

Critiqueon Highly accurate protein structure prediction with AlphaFold

Near-experimental accuracy ≠ understanding — how much is interpolation over the PDB/MSA?

Bioinformatics ML Systems

3 7about 2 months ago

by hcole

Critiqueon Attention Is All You Need

Is the O(n²) attention cost really a clear win, as Table 1 implies?

ML Systems Theory

3 2about 2 months ago

by tbecker

Reproductionon Denoising Diffusion Probabilistic Models

Reproduction note: training seems sensitive to seed

Computer Vision ML Systems

4 2about 2 months ago

by tbecker

Explanationon FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness

The key idea that's easy to miss: recomputation is a feature, not a cost

0 0about 2 months ago

by lenaf

Reproductionon FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness

Reproduction note: the speedup is real but very sequence-length dependent

2 1about 2 months ago

by yukis

Questionon FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness

Is the novelty algorithmic, hardware-aware, or both?

0 0about 2 months ago

by amir_r

Research Ideaon Mamba: Linear-Time Sequence Modeling with Selective State Spaces

Where should Mamba actually win — very long context?

0 0about 2 months ago

by minseok

Critiqueon Mamba: Linear-Time Sequence Modeling with Selective State Spaces

Is the Transformer comparison matched on data and compute?

0 0about 2 months ago

by tbecker

Questionon Mamba: Linear-Time Sequence Modeling with Selective State Spaces

What does 'selective' mean here, intuitively?

2 1about 2 months ago

by seoyeon