Ebtesam @ebtesamdotpy

AI tools for SE research | CS PhD @GeorgeMasonU @INSPIREDLabGMU | Prev @MSFTResearch mason.gmu.edu/~ehaque4 Washington, DC Joined October 2021

Tweets

34
Followers

115
Following

192
Likes

2K

pdawg @prathamgrv

2 months ago

the biggest delusional lie I tell myself is that there’s still time to learn everything

17 10 263 10K 28

As we all know by now, reasoning models often generate longer responses, which raises compute costs. Now, this new paper (arxiv.org/abs/2504.05185) shows that this behavior comes from the RL training process, not from an actual need for long answers for better accuracy. The RL…

33 190 1K 107K 836

Download Image

Tristan T @trirpi

6 months ago

128 1K 10K 675K 731

Download Image

I Am Devloper @iamdevloper

6 months ago

vibe coding, where 2 engineers can now create the tech debt of at least 50 engineers

168 1K 15K 641K 770

Nabeel S. Qureshi @nabeelqu

6 months ago

For the confused, it's actually super easy: - GPT 4.5 is the new Claude 3.6 (aka 3.5) - Claude 3.7 is the new o3-mini-high - Claude Code is the new Cursor - Grok is the new Perplexity - o1 pro is the 'smartest', except for o3, which backs Deep Research Obviously. Keep up.

253 629 11K 1.1M 3K

Hamel Husain @HamelHusain

8 months ago

New post re: Devin (the AI SWE). We couldn't find many reviews of people using it for real tasks, so we went MKBHD mode and put Devin through its paces. We documented our findings here. Would love to know if others have had a different experience. answer.ai/posts/2025-01-…

61 171 2K 570K 1K

Download Image

Diomidis Spinellis @CoolSWEng

11 months ago

Long overdue, a paper finally exposes the Emperor's New “Threats to Validity” Clothes in empirical software engineering research. Even better, it provides suggestions for improving the state of practice.

Prof. Per Runeson @SoftEngResGrp

11 months ago

3 5 36 3K 5

1 4 9 1K 2

Download Image

Jiaxin Pei @jiaxin_pei

11 months ago

It's common to add personas in system prompts, assuming this can help LLMs. However, through analyzing 162 roles x 4 LLMs x 2410 questions, we show that adding a persona mostly has *no* statistically significant difference from the no-persona setting. If there is a difference, it…

Mingqian Zheng @elisazmq_zheng

11 months ago

10 94 431 49K 411

Download Image

1 15 68 9K 37

Upol Ehsan @UpolEhsan

a year ago

Is hallucination in LLMs inevitable even with an idealized model architecture and perfect training data? This work argues YES and offers a formal proof. Let's dig in ⤵ 🧵1/n

15 69 330 59K 496

Download Image

Edward Grefenstette @egrefen

a year ago

Instead, evaluation processes should track the diverse notions of extrinsic utility which are to be found in both everyday usage of our technology today, but also anticipating how people might use technology tomorrow.

1 2 10 1K 1

Dr Meming @Dr_Meming

2 years ago

Heck

14 207 2K 118K 106

Download Image

INSPIRED Lab @ GMU @INSPIREDLabGMU

2 years ago

🚨 Inclusive tech research alert! 🚨 Are you a tech user who identifies as BIPOC (bit.ly/BIPOC_defined)? Or a researcher/practitioner who uses data in your work? Share your experiences in our 20 min. survey→go.gmu.edu/EngagingTheMar… IRBNet #: 1945546-2 #data #tech #trust

1 2 2 896 0

Dr. Amy Lee @minisciencegirl

2 years ago

Never name a manuscript draft "_FINAL"

64 157 1K 155K 36

Dr Meming @Dr_Meming

2 years ago

Academic research: months of experiments and data analysis that ends up being a few sentences in a paper

18 691 6K 315K 267

Download Image

will depue @willdepue

2 years ago

I feel like large language model feels a bit reductive when GPT-2 is in the same class as GPT-4. gigantic language models? enormous language models? big ass language models? Nimitz-class language models? better suggestions needed

71 3 222 32K 11

MIT CSAIL @MIT_CSAIL

2 years ago

Happy birthday to Python creator Guido van Rossum. The open source language was named after comedy troupe Monty Python: bit.ly/2B8R7h6 Image v/Midjourney

8 171 787 56K 46

Download Image

François Chollet @fchollet

2 years ago

When I got started with programming, I debugged using printf() statements. Today, I debug with print() statements. The purpose of debugging is to correct your mental model of what your code does, and no tool can do that for you. The best any tool can do is provide visibility…

MIT CSAIL @MIT_CSAIL

2 years ago

17 235 1K 621K 183

80 212 2K 564K 403

Michael Baym @baym

2 years ago

If there was only one scientific practice I could teach to every scientist regardless of stage or field I think it would be: look at the data. Spot check it. Find a few data points and trace them through to see if they make sense. Look at the raw data. Don't just do analyses.