Ansari Ismail

🚀 Looking to streamline fine-tuning for models like Llama 3, Mistral, and Mixtral? Our latest blog shows how Anyscale simplifies LLMs fine-tuning with Ray: 🔑 Key takeaways: • Scalable fine-tuning with LLM-Forge • Model Deployment with RayLLM Read more: https://lnkd.in/gQwD4A2U

34

[NEW COUPON] 2025 Fine Tuning LLM with Hugging Face Transformers for NLP Course Master Transformer models like Phi2, LLAMA; BERT variants, and distillation for advanced NLP applications on custom data Created by Laxmi Kant | KGP Talkie | 16.5 hours on-demand video course | 1 downloadable resource Fine Tuning LLM Course Overview Welcome to “Mastering Transformer Models and LLM Fine Tuning”, a comprehensive and practical course designed for all levels, from beginners to advanced practitioners in Natural Language Processing (NLP). This course delves deep into the world of Transformer models, fine-tuning techniques, and knowledge distillation, with a special focus on popular BERT variants like Phi2, LLAMA, T5, BERT, DistilBERT, MobileBERT, and TinyBERT. Get Udemy Coupon Code: https://lnkd.in/dr_KGEfw #LLM #Huggingface #NLP 

2

💥💥💥 Align-SLM: Textless Spoken Language Models with Reinforcement Learning from AI Feedback Guan-Ting Lin, Prashanth Gurunath Shivakumar, Aditya Gourav, Yile Gu, Ankur Gandhe, Hung-yi Lee, Ivan Bulyko Abstract While textless Spoken Language Models (SLMs) have shown potential in end-to-end speech-to-speech modeling, they still lag behind text-based Large Language Models (LLMs) in terms of semantic coherence and relevance. This work introduces the Align-SLM framework, which leverages preference optimization inspired by Reinforcement Learning with AI Feedback (RLAIF) to enhance the semantic understanding of SLMs. Our approach generates multiple speech continuations from a given prompt and uses semantic metrics to create preference data for Direct Preference Optimization (DPO). We evaluate the framework using ZeroSpeech 2021 benchmarks for lexical and syntactic modeling, the spoken version of the StoryCloze dataset for semantic coherence, and other speech generation metrics, including the GPT4-o score and human evaluation. Experimental results show that our method achieves state-of-the-art performance for SLMs on most benchmarks, highlighting the importance of preference optimization to improve the semantics of SLMs. 👉 https://lnkd.in/dXBzaibk #machinelearning

4

#Multimodality is quickly moving from an impressive demo feature to a basic expectation in apps harnessing #GenAI. ICYMI, here's a good article by Manoranjan Rajguru on how to apply you RAG solution augmented with Vision capabilities over Hierarchical document structures. https://lnkd.in/gxcDHguw

20

2 Comments

In computational linguistics, researchers are focused on enhancing language models' ability to extract precise information from extensive textual data. Recent studies introduce "retrieval heads," specialized attention mechanisms that selectively focus on critical segments of large texts, improving targeted data retrieval. These heads, integrated into transformer-based models like LLaMA and Yi, significantly enhance accuracy and efficiency, particularly in handling long-context scenarios. The methodology involves thorough experiments across various models, including the Needle-in-a-Haystack test, which embeds specific information to measure precision. Results show models with active retrieval heads outperform those without, with accuracy dropping notably when these heads are masked. The empirical evidence underscores the effectiveness of retrieval heads in improving information retrieval precision and reliability. This research deepens our understanding of attention mechanisms in large-scale text processing, paving the way for more efficient and accurate language models. https://lnkd.in/dNNPvMXw https://lnkd.in/ddbHX6EA Innovation Hacks AI #aiinnovation #generativeai

1

Golden Retrieval - MM-Embed: Universal Multimodal Retrieval with Multimodal LLMs (#NVIDIA, November 2024) Paper: https://lnkd.in/gpmQBsUe Abstract: "State-of-the-art retrieval models typically address a straightforward search scenario, where retrieval tasks are fixed (e.g., finding a passage to answer a specific question) and only a single #modality is supported for both queries and retrieved results. This paper introduces techniques for advancing information retrieval with multimodal large language models (MLLMs), enabling a broader search scenario, termed universal multimodal retrieval, where multiple modalities and diverse retrieval tasks are accommodated. To this end, we first study fine-tuning an MLLM as a bi-encoder retriever on 10 datasets with 16 retrieval tasks. Our empirical results show that the fine-tuned MLLM retriever is capable of understanding challenging queries, composed of both text and image, but underperforms a smaller CLIP retriever in cross-modal retrieval tasks due to modality bias from #MLLMs. To address the issue, we propose modality-aware hard negative mining to mitigate the modality bias exhibited by MLLM retrievers. Second, we propose to continually fine-tune the universal multimodal retriever to enhance its text retrieval capability while maintaining multimodal retrieval capability. As a result, our model, MM-Embed, achieves state-of-the-art performance on the multimodal retrieval benchmark M-BEIR, which spans multiple domains and tasks, while also surpassing the state-of-the-art text retrieval model, NV-Embed-v1, on MTEB retrieval benchmark. Finally, we explore to prompt the off-the-shelf MLLMs as the zero-shot rerankers to refine the ranking of the candidates from the multimodal retriever. We find that through #prompt-and-reranking, MLLMs can further improve multimodal retrieval when the user queries (e.g., text-image composed queries) are more complex and challenging to understand. These findings also pave the way to advance universal multimodal retrieval in the future." #genai #retriever #multimodal #search #information Zachary Huhn Summer Crenshaw Scott Phillips Lohith Srikanth Pentapalli Kelly Cohen

7

3 Comments

Wondering how various parameters affect Retrieval-Augmented Generation (RAG) performance in practice ⁉️ New blog episode presents hands-on exploration of factors like chunk size, overlap, and context size, examining their impact on the LLM’s ability to answer questions. 🛠️ It also delves into error analysis and the cost breakdown of running the full pipeline, including token usage and pricing considerations.💡 Read now and thank Luka Panić later for nailing it again! 🏆🌟➡️ bit.ly/3A3jSgC #PixionAiBlogSeries #AI #LLM #RAG

7

vGPU, MIG, & Time-Slicing! 3 ways to optimize the #GPU resources to meet the dynamic demands of computational tasks💻 Sameer Kulkarni explains & compares these methods in the blog post to help you select the right GPU optimization for your models👇 https://lnkd.in/gFQPX4b4

7

New Best Practice on CPT 👏🏻 Continued Pre-Training (CPT) refers to the cost effective alternative to pre-training for a new, specific domain (eg. a medical domain) that was not well represented in a model’s pre-training corpus. In this blog, we outline the entire CPT process for: Showing how to optimize hyperparameters. Measuring the effect of different datasets. Developing heuristics for mixing datasets. Mitigating forgetting.

14

🤖 Why JEPA Proves LLMs are Inefficient for Human-Level Reasoning 🤖 Join Karthikeyan VK from Peninsular Research Operation for a compelling session that explores the limitations of Large Language Models (LLMs) and introduces JEPA (Joint Embedding Predictive Architecture) as a more efficient alternative for achieving human-level reasoning. Learn why LLMs, despite their advanced language skills, struggle with contextual understanding and complex decision-making and see how JEPA offers a framework that more closely mirrors human cognitive processes. Through real-world examples, discover the future potential of JEPA in applications where deep understanding and adaptability are key. Register Now: https://techxconf.com #AI #JEPA #LLMLimitations #HumanCognitiveAI #TechXConf2024 #FutureOfAI #TechXConf

20

𝗪𝗶𝗹𝗹 𝗔𝗜 𝗛𝗶𝘁 𝗮 𝗗𝗮𝘁𝗮 𝗪𝗮𝗹𝗹 𝗶𝗻 𝟮𝟬𝟯𝟬? 🆘 Very interesting analysis from epoch.ai about data corpus available for training large language models and whether we will hit the ceiling soon. https://lnkd.in/gD78f9DW The authors estimate that all available data will be used by ~2030, but they're also the first to admit that by then we will have ways to address "the shortage" (such as synthetic data, data efficiency improvements, use of other modalities). What I found more interesting though is why the latest predictions differ from their 2022 paper, where they speculated that all the data will be used by 2024: ☑ Web data proves more valuable than expected (on par with scientific papers) ☑ Models can "reuse" datasets multiple times 2030 is right around the corner, so we will see a lot of innovation on this front and creative ideas on how to 'recycle' and make best use of the data we already have. These estimates also don't take into account proprietary data owned by individual companies. In this No priors episode (https://lnkd.in/g2BkTJBz) there is a mention of JPMorgan owning 150PB of proprietary data, which is at least 150x of what was used to train GPT-4. JPMorgan is monetizing this data today (https://lnkd.in/gXjE92qx), but it gets exponentially more valuable in a world where data is a finite (and extremely sought after) resource.

12

Benchmarking Text Generation Inference (TGI) is essential to optimize Large Language Model (LLM) deployments. The TGI benchmarking tool from Hugging Face helps users understand trade-offs between latency and throughput, and customize configurations to meet specific needs. This tool simplifies tuning LLMs for various use cases, enhancing performance while managing costs effectively. Whether optimizing for user experience or maximizing throughput, the TGI tool offers valuable insights. Learn more about the TGI benchmarking tool and its applications in improving LLM deployments. Full Article: https://lnkd.in/dZXHb8wT

4

As many go down the "agentic path", we're choosing a different path.....graph path traversals! 🔵->🟨->🟢 Graph path traversals use vector similarity and/or relationships of your choosing to walk a graph and enable LLMs to explain complex concepts and relationships. This example walks a path and automatically generates an explanation of the network in the form of a short article. Paths can be set directly (i.e. Roman Empire -> Reasons for collapse) or inferred from a query (Tell me the reasons why the Roman Empire collapsed). Learn more here: https://lnkd.in/evdB5HgN

8

Foundational models, or large language models (LLMs), train at the scale of the entire internet to acquire the vocabulary, syntax, and narrative structure of human language. Fine-tuning involves using a smaller, domain-specific dataset to grant additional specificity to a pre-existing foundational model that has already acquired patterns and features from massive underlying training datasets. By utilizing the existing knowledge embedded in a pre-trained language model, we unlock high-performance language understanding, but fine-tuning helps to get the right answer - not just an answer that sounds right - with a minimum of hallucinations or incorrect info. In healthcare, out-of-the-box, foundational models are often too generic to sufficiently answer the precision of a healthcare query. Foundational language models are trained on a massive scale of data, which includes jokes, incorrect information, misunderstandings, and misinformation, along with the sum total of human knowledge. The problem is, if you ask a large language model to give you factually correct information when it has been trained on BOTH accurate and inaccurate information - you’re playing a bit of a dangerous game. Enter fine-tuning language models. In the old days, we used to build a single model for a single task - for example, classifying procedures into evidence-based or not as laid out by clinical practice guidelines, for patients with particular disease markers. Each disease would have its own model, which took a long time to build and didn’t scale particularly well, and we struggled with the sparsity and inherent high-dimensionality of healthcare data. By fine-tuning LLMs, however, we're able to specialize those foundational models to achieve greater domain specificity and specialization, leverage the unique predictive and planning capabilities of LLMs, and cut down on the incorrect information that such models inevitably produce, while still preserving the underlying language prowess of the LLM. If improved accuracy and specificity wasn't enough of a selling point, there's another notable upside to fine-tuning foundational models for use in a clinical context. In terms of computational efficiency, using LLMs out of the box on specific tasks can require massive compute (and spend) that amounts to computational overkill for the problem. Some LLMs charge per query, some by compute load or time - either way, as a business running LLMs in a production environment, minimizing the number of LLM calls means optimizing your compute expenditures, too. When LLMs fail at scale in healthcare, it's usually along the fault line of failure to specify adequately. Fine-tuning large language models along domain-specific lines is one way to effectively productize genAI for healthcare use cases. #genAI #LLM #ai #healthcareAI #healthtech #largelanguagemodels #generativeAI #chatGPT

61

2 Comments

Check out this tutorial authored by senior Generative AI engineer Bhavishya Pandit on how you can use LLMs to generate pre-processed, high-quality data for classic ML models: https://ibm.co/3Vjvt2u Through code examples and practical guidance, learn how LLM-driven data generation can enhance ML model performance by reducing overfitting risks, preserving data quality, and expanding the capabilities of classic ML applications. #AI #ML #GenAI

34

6 Comments

We just released a blog on whether the use of multimodal feature transfer can detect deception (phenomena like sarcasm, irony, and condescension) in conversation. The short answer... Yes. This particular write-up profiles a subset of our work at Symbl.ai on multimodal large language models, and is research led by one of our summer interns, Benjamin R. (from Prof. Larry Heck's AVA Lab at Georgia Institute of Technology). We have a paper currently under submission on this work, and we can't wait to share the full details with all of you soon. Blog Link: https://lnkd.in/g6m_wFDi #multimodal #featuretransfer #LLMs #deception #conversation #sarcasm #irony #language #AI #ML #NLP #LargeLanguageModels #SummerInternship #ResearchPapers #ACL #Llama2 #SymblDotAI

16