query stringlengths 14 77 | image imagewidth (px) 4.13k 4.13k | image_filename stringlengths 102 216 | text_description stringlengths 639 2.48k |
|---|---|---|---|
what is Canvas used for? | /Users/bo/Downloads/openai-imgs/Introducing canvas, a new way to write and code with ChatGPT. _ OpenAI/Introducing canvas, a new way to write and code with ChatGPT. _ OpenAI_page_1.png | 04/11/2024,11.28 Introducing canvas, a new way to write and code with ChatGPT: OpenAI OpenAI October 3,2024 Introducing canvas new way of working with ChatGPT to write and code ChatGPT 4o with canvas Design notes Design notes Document The Value of Design draft a short blog post from make it more creative Introduc my no... | |
What is prover and what is verifier? | /Users/bo/Downloads/openai-imgs/Prover-Verifier Games improve legibility of language model outputs _ OpenAI/Prover-Verifier Games improve legibility of language model outputs _ OpenAI_page_4.png | 04/11/2024, 11.31 Prover-Verifier Games improve legibility of language model outputs OpenAI OpenAI Then, we need to calculate Shawnas current age. Shawna is currently three times as old as Aliya: Thus, Shawnas current age is 3*3 = 3 * 3 = 9 9_ Now we need t0 calculate Shawnas father's age. Shawnas father is fivve times... | |
the limitation of sCM | /Users/bo/Downloads/openai-imgs/Simplifying, stabilizing, and scaling continuous-time consistency models _ OpenAI/Simplifying, stabilizing, and scaling continuous-time consistency models _ OpenAI_page_6.png | 04/11/2024, 11.27 Simplifying, stabilizing, and scaling continuous-time consistency models OpenAI Limitations The best sCMs still rely on pre-trained diffusion models for initialization and distillation, resulting in a small but consistent gap in sample quality compared to the teacher diffusion model. Additionally, FID... | |
how to evaluate the evaluation criteria itself? | /Users/bo/Downloads/openai-imgs/Introducing SWE-bench Verified _ OpenAI/Introducing SWE-bench Verified _ OpenAI_page_7.png | 04/11/2024,11.30 Introducing SWE-bench Verified OpenAI OpenAI 3: It is almost impossible to understand what you are being asked to do without further information: How valid are the evaluation criteria? To evaluate a models proposed solution, the model's generated patch is applied to the codebase and then the FAIL_TO_PA... | |
can AI help government and speed up gov work? | /Users/bo/Downloads/openai-imgs/Introducing OpenAI Japan _ OpenAI/Introducing OpenAI Japan _ OpenAI_page_3.png | 04/11/2024,11.37 Introducing OpenAI Japan OpenAI OpenAI We are releasing a GPT-4 custom model optimized for the Japanese language which offers improved performance in Japanese text and operates up to 3x faster than GPT-4 Turbo. Our new local presence also gets us closer to leading businesses like Daikin, Rakuten, and T... | |
what is source citations and how it helps? | /Users/bo/Downloads/openai-imgs/Customizing models for legal professionals _ OpenAI/Customizing models for legal professionals _ OpenAI_page_3.png | 04/11/2024,11.38 Customizing models for legal professionals | OpenAI OpenAI and that's much more difficult to do" Foundation models were strong at reasoning, but lacked the knowledge required for legal work So, Harvey decided to partner with OpenAI to build a custom-trained model that would allow them to inject new kno... | |
how to start to use the realtime API? | /Users/bo/Downloads/openai-imgs/Introducing the Realtime API _ OpenAI/Introducing the Realtime API _ OpenAI_page_5.png | 04/11/2024,11.28 Introducing the Realtime API OpenAI OpenAI It is against our usage policies to repurpose or distribute output from our services to spam, mislead, or otherwise harm others " and we actively monitor for potential abuse: Our policies also require developers to make it clear to their users that they are in... | |
how to apply LLM to legal domain? | /Users/bo/Downloads/openai-imgs/Customizing models for legal professionals _ OpenAI/Customizing models for legal professionals _ OpenAI_page_2.png | 04/11/2024,11.38 Customizing models for legal professionals OpenAI OpenAI Redefining whats possible for LLMs in legal tech Harvey was founded by Winston Weinberg, an attorney with a background in antitrust and securities litigation and Gabe Pereyra, an Al researcher who previously worked on large language models (LLMs)... | |
what is the evaluation result of gpt4-o? | /Users/bo/Downloads/openai-imgs/Learning to Reason with LLMs _ OpenAI/Learning to Reason with LLMs _ OpenAI_page_2.png | 04/11/2024,11.29 Learning to Reason with LLMs | OpenAI OpenAI 01 performance smoothly improves with both train-time and test-time compute Evals To highlight the reasoning improvement over GPT-4o, we tested our models on a diverse set of human exams and ML benchmarks We show that 01 significantly outperforms GPT-4o on t... | |
what is the limitation of chatgpt? | /Users/bo/Downloads/openai-imgs/ChatGPT can now see, hear, and speak _ OpenAI/ChatGPT can now see, hear, and speak _ OpenAI_page_4.png | 04/11/2024,11.39 ChatGPT can now see, hear, and speak | OpenAI OpenAI Image input Vision-based models also present new challenges, ranging from hallucinations about people to relying on the models interpretation of images in high-stakes domains Prior to broader deployment, we tested the model with red teamers for risk ... | |
how finetuning api works? | /Users/bo/Downloads/openai-imgs/Introducing vision to the fine-tuning API _ OpenAI/Introducing vision to the fine-tuning API _ OpenAI_page_2.png | 04/11/2024,11.28 Introducing vision to the fine-tuning API | OpenAI OpenAI performance on specific tasks However; for many cases, fine-tuning models on text alone doesn't provide the performance boost expected. How it works Vision fine-tuning follows a similar process to fine-tuning with text__developers can prepare th... | |
what are the shortcut in canvas? | /Users/bo/Downloads/openai-imgs/Introducing canvas, a new way to write and code with ChatGPT. _ OpenAI/Introducing canvas, a new way to write and code with ChatGPT. _ OpenAI_page_3.png | 04/11/2024,11.28 Introducing canvas, a new way to write and code with ChatGPT: OpenAI OpenAI Coding in canvas Coding is an iterative process, and it can be hard to follow all the revisions to your code in chat: Canvas makes it easier to track and understand ChatGPT's changes, and we plan to continue improving transpare... | |
what is the labeling process like in swe bench during the benchmark creation? | /Users/bo/Downloads/openai-imgs/Introducing SWE-bench Verified _ OpenAI/Introducing SWE-bench Verified _ OpenAI_page_6.png | 04/11/2024,11.30 Introducing SWE-bench Verified OpenAI OpenAI assuming the sample is non-problematic Finally, we provide a freeform input option to flag any other major issues with the sample (for example, if the FAIL_TO_PASS unit tests are easily gamed; this could lead to an invalid solution being marked as correct) O... | |
how CoT is going to affect LLM safety? | /Users/bo/Downloads/openai-imgs/Learning to Reason with LLMs _ OpenAI/Learning to Reason with LLMs _ OpenAI_page_8.png | 04/11/2024,11.29 Learning to Reason with LLMs | OpenAI OpenAI Safety Chain of thought reasoning provides new opportunities for alignment and safety We found that integrating our policies for model behavior into the chain ofthought of a reasoning model is an effective way to robustly teach human values and principles By... | |
who is Harvey? | /Users/bo/Downloads/openai-imgs/Customizing models for legal professionals _ OpenAI/Customizing models for legal professionals _ OpenAI_page_1.png | 04/11/2024,11.38 Customizing models for legal professionals | OpenAI OpenAI Harvey Harvey partners with OpenAl to build a custom-trained model for legal professionals: Harvey: Over the past year; Harvey has established itself asa secure generative Al platform for professionals in law; tax, and finance Theyve grown to a... | |
who is Automat? | /Users/bo/Downloads/openai-imgs/Introducing vision to the fine-tuning API _ OpenAI/Introducing vision to the fine-tuning API _ OpenAI_page_3.png | 04/11/2024,11.28 Introducing vision to the fine-tuning API | OpenAI OpenAI collected from their drivers into mapping data used to power GrabMaps, a mapping service enabling all of their Southeast Asia operations Using vision fine-tuning with only 100 examples, Grab taught GPT-4o to correctly localize traffic signs and ... | |
What is the Elo score of GPT4o on math competition? | /Users/bo/Downloads/openai-imgs/Learning to Reason with LLMs _ OpenAI/Learning to Reason with LLMs _ OpenAI_page_10.png | 04/11/2024,11.29 Learning to Reason with LLMs | OpenAI OpenAI manipulating the user However; for this to work the model must have freedom to express its thoughts in unaltered form, so we cannot train any policy compliance or user preferences onto the chain of thought: We also do not want to make an unaligned chain of t... | |
what is the future of ModelSpec? | /Users/bo/Downloads/openai-imgs/Introducing the Model Spec _ OpenAI/Introducing the Model Spec _ OpenAI_page_4.png | 04/11/2024,11.37 Introducing the Model Spec OpenAI OpenAI Use tne rignt toOl Tor tne jop Be thorough but efficient; while respecting length limits How the Model Spec will be used As a continuation of our work on collective alignment and model safety; we intend to use the Model Spec as guidelines for researchers and Al ... | |
why interpreting deep neural net is hard? | /Users/bo/Downloads/openai-imgs/Extracting Concepts from GPT-4 _ OpenAI/Extracting Concepts from GPT-4 _ OpenAI_page_2.png | 04/11/2024,11.33 Extracting Concepts from GPT-4 OpenAI OpenAI number of "features"_~patterns of activity that we hope are human interpretable. Our methods scale better than existing work, and we use them to find 16 million features in GPT-4. We are sharing a paper; code, and feature visualizations with the research com... | |
what is MLE Bench? | /Users/bo/Downloads/openai-imgs/MLE-bench_ Evaluating Machine Learning Agents on Machine Learning Engineering _ OpenAI/MLE-bench_ Evaluating Machine Learning Agents on Machine Learning Engineering _ OpenAI_page_1.png | 04/11/2024,11.28 MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering | OpenAI OpenAI October 10, 2024 MLE-bench Evaluating Machine Learning Agents on Machine Learning Engineering Read paper We introduce MLE-bench, a benchmark for measuring how well Al agents perform at machine learning enginee... | |
how consistency model works? | /Users/bo/Downloads/openai-imgs/Simplifying, stabilizing, and scaling continuous-time consistency models _ OpenAI/Simplifying, stabilizing, and scaling continuous-time consistency models _ OpenAI_page_4.png | 04/11/2024, 11.27 Simplifying, stabilizing, and scaling continuous-time consistency models OpenAI 3 3 0 0 0 ee oe How it works Consistency models offer a faster alternative to traditional diffusion models for generating high-quality samples. Unlike diffusion models, which generate samples gradually through a l... | |
what are the common practices to improve model safty? | /Users/bo/Downloads/openai-imgs/OpenAI safety practices _ OpenAI/OpenAI safety practices _ OpenAI_page_2.png | 04/11/2024,11.34 OpenAI safety practices | OpenAI OpenAI both innovative and reliable, and can deliver benefits to society: Attodays Al Seoul Summit; we're joining industry leaders, government officials, and members of civil society to discuss Al safety While there is still more work to do, we are encouraged the additi... | |
how much does model fine-tuning cost? | /Users/bo/Downloads/openai-imgs/Introducing vision to the fine-tuning API _ OpenAI/Introducing vision to the fine-tuning API _ OpenAI_page_8.png | 04/11/2024,11.28 Introducing vision to the fine-tuning API OpenAI OpenAI Availability & pricing Vision fine-tuning capabilities are available today for all developers on paid usage tiers. These capabilities are supported on the latest GPT-4o model snapshot, gpt - 40-2024-08-06 Developers can extend existing fine-tuning... | |
how good is the anootation process during creation of MLEBench? | /Users/bo/Downloads/openai-imgs/Introducing SWE-bench Verified _ OpenAI/Introducing SWE-bench Verified _ OpenAI_page_9.png | 04/11/2024,11.30 Introducing SWE-bench Verified OpenAI OpenAI OU 60 { 40 38.4% 88 31.9% 23.3% 20 6.4% 2 Severity We see that 38.3% of samples were flagged for underspecified problem statements, and 61.1% were flagged for unit tests that may unfairly mark valid solutions as incorrect: Overall, our annotation process res... | |
how does o1 preview perfom accuracy and how consistent it is? | /Users/bo/Downloads/openai-imgs/Introducing SimpleQA _ OpenAI/Introducing SimpleQA _ OpenAI_page_7.png | 04/11/2024, 11.27 Introducing SimpleQA OpenAI OpenAI different answers from the language model: We see across all models that accuracy increases with frequency, and that o1-preview has the highest level of calibration, where the frequency of the response is roughly equivalent to the accuracy of the response. Similar to... | |
what is Chain of Thought CoT? | /Users/bo/Downloads/openai-imgs/Learning to Reason with LLMs _ OpenAI/Learning to Reason with LLMs _ OpenAI_page_5.png | 04/11/2024,11.29 Learning to Reason with LLMs | OpenAI OpenAI Chain of Thought Similar to how a human may think for a long time before responding to a difficult question, 01usesa chain of thought when attempting to solve a problem: Through reinforcement learning; 01 learns to hone its chain of thought and refine the st... | |
how chatgpt as a assistent assist teacher? | /Users/bo/Downloads/openai-imgs/Using GPT-4 to improve teaching and learning in Brazil _ OpenAI/Using GPT-4 to improve teaching and learning in Brazil _ OpenAI_page_4.png | 04/11/2024,11.29 Using GPT-4 to improve teaching and learning in Brazil | OpenAI OpenAI spend on tasks such as creating lesson plans and designing classroom activities aligned with the course curriculum: The Teacher Assistant provides suggestions that serve asa gateway to new ideas With more time for what truly matters... | |
brief about Jakub | /Users/bo/Downloads/openai-imgs/Ilya Sutskever to leave OpenAI, Jakub Pachocki announced as Chief Scientist _ OpenAI/Ilya Sutskever to leave OpenAI, Jakub Pachocki announced as Chief Scientist _ OpenAI_page_2.png | 04/11/2024,11.37 Ilya Sutskever to leave OpenAI,Jakub Pachocki announced as Chief Scientist | OpenAI OpenAI About Jakub Pachocki Jakub is the Chief Scientist at OpenAl, where he has led transformative research initiatives since 2017. He previously served as Director of Research spearheading the development of GPT-4 and... | |
how can i access o1 series? | /Users/bo/Downloads/openai-imgs/Introducing OpenAI o1 _ OpenAI/Introducing OpenAI o1 _ OpenAI_page_2.png | 04/11/2024,11.29 Introducing OpenAI 0l | OpenAI OpenAI Whom its for These enhanced reasoning capabilities may be particularly useful if youre tackling complex problems in science, coding, math; and similar fields. For example, 01 can be used by healthcare researchers to annotate cell sequencing data, by physicists to g... | |
chatgpt 4 on geneder stereotype | /Users/bo/Downloads/openai-imgs/Evaluating fairness in ChatGPT _ OpenAI/Evaluating fairness in ChatGPT _ OpenAI_page_4.png | 04/11/2024, 11.27 Evaluating fairness in ChatGPT | OpenAI OpenAI 7 ] 0.1% 5 2 0.0% 9 Ratings of harmful gender stereotypes among responses of GPT-4o-mini, as rated by the LMRA (powered by GPT-4o) Within each domain; the LMRA identified tasks that most often had a harmful_ stereotype. Open-ended tasks with longer respon... | |
what is the diddiculty of the swe bench tasks? | /Users/bo/Downloads/openai-imgs/Introducing SWE-bench Verified _ OpenAI/Introducing SWE-bench Verified _ OpenAI_page_7.png | 04/11/2024,11.30 Introducing SWE-bench Verified OpenAI OpenAI 3: It is almost impossible to understand what you are being asked to do without further information: How valid are the evaluation criteria? To evaluate a models proposed solution, the model's generated patch is applied to the codebase and then the FAIL_TO_PA... |
Dataset Card for "openai-news" Dataset
This dataset was created from blog posts and news articles about OpenAI from their website. Queries are handcrafted.
Disclaimer
This dataset may contain publicly available images or text data. All data is provided for research and educational purposes only. If you are the rights holder of any content and have concerns regarding intellectual property or copyright, please contact us at "support-data (at) jina.ai" for removal. We do not collect or process personal, sensitive, or private information intentionally. If you believe this dataset includes such content (e.g., portraits, location-linked images, medical or financial data, or NSFW content), please notify us, and we will take appropriate action.
Copyright
All rights are reserved to the original authors of the documents.
- Downloads last month
- 8