<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Abhinav Prasad Yasaswi on Medium]]></title>
        <description><![CDATA[Stories by Abhinav Prasad Yasaswi on Medium]]></description>
        <link>https://medium.com/@abhinavp_41237?source=rss-5ffa75f9ef71------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*F8YMRqLBRe1JQJMO32xEEw.png</url>
            <title>Stories by Abhinav Prasad Yasaswi on Medium</title>
            <link>https://medium.com/@abhinavp_41237?source=rss-5ffa75f9ef71------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 22:46:05 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@abhinavp_41237/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[LLM Benchmarks? What are they? Who are they?]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://aiadvances.org/llm-benchmarks-what-are-they-who-are-they-154f6b964656?source=rss-5ffa75f9ef71------2"><img src="https://cdn-images-1.medium.com/max/1776/1*xregbhgmTqBYEwJbqejmkg.png" width="1776"></a></p><p class="medium-feed-snippet">Ever wondered what these benchmarks that companies like OpenAI, Google, Anthropic, and xAI claim their LLM models do great on? Read on&#x2026;</p><p class="medium-feed-link"><a href="https://aiadvances.org/llm-benchmarks-what-are-they-who-are-they-154f6b964656?source=rss-5ffa75f9ef71------2">Continue reading on AI Advances »</a></p></div>]]></description>
            <link>https://aiadvances.org/llm-benchmarks-what-are-they-who-are-they-154f6b964656?source=rss-5ffa75f9ef71------2</link>
            <guid isPermaLink="false">https://medium.com/p/154f6b964656</guid>
            <category><![CDATA[openai]]></category>
            <category><![CDATA[large-language-models]]></category>
            <category><![CDATA[chatgpt]]></category>
            <category><![CDATA[benchmark]]></category>
            <category><![CDATA[google]]></category>
            <dc:creator><![CDATA[Abhinav Prasad Yasaswi]]></dc:creator>
            <pubDate>Thu, 26 Sep 2024 20:05:22 GMT</pubDate>
            <atom:updated>2024-09-26T20:05:22.245Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[ChatGPT’s advanced voice mode is here! My first impressions…]]></title>
            <link>https://aiadvances.org/chatgpts-advanced-voice-mode-is-here-my-first-impressions-df92a1b36b52?source=rss-5ffa75f9ef71------2</link>
            <guid isPermaLink="false">https://medium.com/p/df92a1b36b52</guid>
            <category><![CDATA[openai]]></category>
            <category><![CDATA[advanced-voice-mode]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[chatgpt]]></category>
            <category><![CDATA[mobile-apps]]></category>
            <dc:creator><![CDATA[Abhinav Prasad Yasaswi]]></dc:creator>
            <pubDate>Thu, 26 Sep 2024 09:20:41 GMT</pubDate>
            <atom:updated>2024-09-26T09:20:41.688Z</atom:updated>
            <content:encoded><![CDATA[<h3>Say hello to advanced voice mode on your ChatGPT mobile app now! (Literally)</h3><p>I began trying out the new voice mode on my Android phone immediately after Sam Altman announced that this feature had been rolled out late midnight at around 1 a.m. MST. Let’s see what’s new, what’s interesting and what’s stupid from this new feature release!</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/720/1*hEEo7-wyroQqzjRZV1D4Iw.jpeg" /><figcaption>Figure 1: The circular icon with a white background and black vertical bars in the center, resembling an audio waveform or a sound equalizer is the button you will have to click to access this feature.</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/720/1*OiNpO4Nfjx1yboIzQo_7bw.jpeg" /><figcaption>Figure 2: You will then be greeted with an interface introducing the new feature.</figcaption></figure><p>Sam Altman had <a href="https://x.com/sama/status/1838864011321872407">announced</a> that OpenAI has completed rolling out its new advanced voice mode on their mobile app. I opened up my app and immediately and updated it so I could access the latest feature that was out. After the update was complete I notice this circular icon with a white background next to my text input box (see Figure 1). I click it with no hesitation and was greeted with an introductory interface describing what this new feature is (see Figure 2). I immediately press on continue and a voice starts talking to me in a calm and soothing voice, asking me what’s on my mind?</p><p>Spruce did come across very composed and gentlemanly and I was definitely impressed. But, I still wanted to find that Scarlett Johanson voice that was all the buzz a few months back so I tried switching over to find *that* voice. But, it seemed like OpenAI preferred to not keep that voice in this release following so much controversy.</p><blockquote>There seems to be <strong>nine</strong> different voices included with the new feature. I wasn’t able to swipe through voices to change over to a different voice on the app initially but then I realized that I had to <strong>swipe right over the name </strong>and <strong>not the circular image</strong>.</blockquote><figure><img alt="" src="https://cdn-images-1.medium.com/max/720/1*Z9tcYm-1alsxK9JG_m7Msw.jpeg" /><figcaption>Figure 3: You will have to swipe right over the name/title of the voice to swipe to the next voice.</figcaption></figure><p>The available voices are as follows: <strong>Spruce, Breeze, Vale, Sol, Ember, Arbor, Maple, Juniper, and Cove</strong>. I noticed they all seem to have either an American or a British accent to start off, but you can explicitly ask the assistant feature to change its accent to your liking during a conversation.</p><p>When I started conversing with it, I started with their first available voice, Spruce, I noticed that this feature is in no way like any of the robotic voice features that we experienced before (I’m talking about Alexa, Google Assistant, Siri, etc.) When I started talking to it, I noticed that it was clear and articulate and was very calm in its tone. The first thing I asked was for it to help me make breakfast, and it suggested starting with avocado toast, and it went on with describing a recipe. When I mentioned I didn’t have avocados, it then suggested that I prepare a veggie omelet and similarly started off eliciting a recipe for it. <strong><em>I interrupted mid-conversation</em></strong>, stating that I didn’t have any vegetables either, and it stopped immediately once it heard my voice and switched from speaking to listening. It quickly understood what I said and immediately, with minimal pause, suggested that I prepare scrambled eggs or a plain omelet instead.</p><blockquote>It was astounding to me how it was able to stop mid-conversation when I interrupted it and went on immediately with a newer response, just like a normal human being would.</blockquote><blockquote>Also, the pace at which it was able to understand my queries, provide answers, and engage in conversation without any latency is incredible.</blockquote><p>In a YouTube video by <a href="https://youtu.be/tuvn-L38Mws?si=qTfUYtaYUeE8Js6k">AI Search</a> the voice feature was asked to count to 100 very quickly and without pausing, but it took a short pause to breath after 29, when the user identified this, he asked the AI to do it without taking a breather and the voice replied that it wished it could, and insisted that it needed to take short breaks between numbers. This opens up questions like, what if the voice is artificially trained to sound human (i.e., what if it’s just top-layer training to sound human?). But if you force it into circumstances where it would make it sound robotic it would immediately deny getting into that situation, because it isn’t a difficult task to count to 100 without taking a breather.’</p><p>You can also see that in this video (<a href="https://youtu.be/tuvn-L38Mws?si=qTfUYtaYUeE8Js6k">AI Search’s YT Video</a>) the voice is able to talk in multiple languages and is able to differentiate between different dialects. I tried talking to it in Hindi and asked it to talk to me in a particular accent that is of someone from the state of Haryana (I could verify with my roommate who is from there that it gets the accent right), and it was able to keep up a full conversation with him in Haryanvi accented Hindi.</p><p>It’s important to note that OpenAI showcased certain video features (in it’s social media posts) in which it was able to look into your room and assess your surroundings. It seems like we will still have to wait for more time for that feature to be rolled out.</p><p>Also, just another observation, I did not understand the motive behind simply deploying this only to their mobile app and not the desktop version. Is it to encourage their app adoption?</p><p>Also, what’s next for OpenAI with these voice features? Will they start deploying home assistants like Google Home or Alexa Echo?</p><p>For OpenAI in general, 2024 has been a top year with incredible announcements like Sora, this new voice feature, and the o1 model. It’s almost as if they are getting closer and closer to Artificial General Intelligence before any of the other companies would. Sora is yet to be deployed to the public for full use. The other features were almost as good as promised and I think we can hope that Sora would be as good as promised.</p><h3>Bonus Finds</h3><p>I have also been surfing on X.com (Twitter) to see other people’s experience to using the new advanced voice feature.</p><p>This user states that the advanced voice mode can sing, hum, recognize and imitate other voices, and even flirt — but it seems to be instructed not to <a href="https://x.com/0xAyush1/status/1838911712780816527">https://x.com/0xAyush1/status/1838911712780816527</a></p><p>This user used this feature to commentate on a football (soccer) match. <a href="https://x.com/ThePyCoach/status/1838965517102596559">https://x.com/ThePyCoach/status/1838965517102596559</a></p><p>There also seems to be a daily limit currently to users <a href="https://x.com/argofowl/status/1838684990051176704">https://x.com/argofowl/status/1838684990051176704</a></p><p>@JStaatsCPA explores the incredible voice modulation features with accent changes, tone changes, and a lot of other interesting voice features: <a href="https://x.com/JStaatsCPA/status/1838984688917954621">https://x.com/JStaatsCPA/status/1838984688917954621</a></p><p>A user manipulates the voice assistant to force it to talk like it’s singing in the opera. <a href="https://x.com/stevelizcano/status/1838847973272305676">https://x.com/stevelizcano/status/1838847973272305676</a></p><p>Expressive powers of ChatGPT: <a href="https://x.com/cyberyogi_/status/1838885166132920482">https://x.com/cyberyogi_/status/1838885166132920482</a></p><p>Prepare for your mock interview: <a href="https://x.com/WevolverApp/status/1838775509687128246">https://x.com/WevolverApp/status/1838775509687128246</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=df92a1b36b52" width="1" height="1" alt=""><hr><p><a href="https://aiadvances.org/chatgpts-advanced-voice-mode-is-here-my-first-impressions-df92a1b36b52">ChatGPT’s advanced voice mode is here! My first impressions…</a> was originally published in <a href="https://aiadvances.org/">AI Advances</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[OpenAI o1: Is This the Enigmatic Force That Will Reshape Every Knowledge Sector We Know?]]></title>
            <link>https://medium.com/data-science/openai-o1-the-enigmatic-force-that-will-reshape-every-knowledge-sector-that-we-know-of-or-99396d641fff?source=rss-5ffa75f9ef71------2</link>
            <guid isPermaLink="false">https://medium.com/p/99396d641fff</guid>
            <category><![CDATA[openai]]></category>
            <category><![CDATA[chatgpt]]></category>
            <category><![CDATA[large-language-models]]></category>
            <category><![CDATA[openai-o1]]></category>
            <category><![CDATA[chain-of-thought]]></category>
            <dc:creator><![CDATA[Abhinav Prasad Yasaswi]]></dc:creator>
            <pubDate>Mon, 16 Sep 2024 05:35:08 GMT</pubDate>
            <atom:updated>2024-09-19T17:01:27.780Z</atom:updated>
            <content:encoded><![CDATA[<h4>My first encounters with the o1 model</h4><figure><img alt="An image generated by DALL-E with a prompt the same as the blog title." src="https://cdn-images-1.medium.com/max/1024/1*Bh01W8elWj5paq8T_3a1tQ.jpeg" /><figcaption>An image generated by DALL-E with a prompt precisely the same as the blog title.</figcaption></figure><h3>My first encounters with the o1 model</h3><p>On the 12th of September at 10:00 a.m., I was in the class “<strong>Frontier Topics in Generative AI</strong>,” a graduate-level course at Arizona State University. A day before this, on the 11th of September, I submitted a team assignment that involved trying to identify flaws and erroneous outputs generated by GPT-4 (essentially trying to prompt GPT-4 to see if it makes mistakes on trivial questions or high-school-level reasoning questions) as part of another graduate-level class “<strong>Topics in Natural Language Processing.”</strong> We identified several trivial mistakes that GPT-4 made, one of them being <em>unable to count the number of r’s in the word strawberry</em>. Before submitting this assignment, I researched several peer-reviewed papers on the internet that identified where and why GPT -4 made mistakes and how you could rectify them. Most of the documents I came across identified two main domains where GPT-4 erred, and they dealt with <strong>planning and reasoning</strong>.</p><p>This paper¹ (although almost a year old) goes in depth through several cases where GPT-4 fails to answer trivial questions that involve simple counting, simple arithmetic, elementary logic, and even common sense. The paper¹ reasons that these questions require some level of reasoning and that because GPT-4 is utterly incapable of reasoning, it almost always gets these questions wrong. The author also states that reasoning is a (very) computationally hard problem. Although GPT-4 is very compute-intensive, its <strong>compute-intensive nature is not geared towards involving reasoning</strong> in solving the questions that it’s prompted with. Several other papers echo this notion of GPT-4 being unable to reason or plan²³.</p><p>Well, let’s get back to the 12th of September. My class ends at around 10:15 a.m., and I come back straight home from class and open up YouTube on my phone as I dig into my morning brunch. The first recommendation on my YouTube homepage was a video from OpenAI announcing the release of GPT-o1 named “<a href="https://youtu.be/3k89FMJhZ00?feature=shared">Building OpenAI o1</a>”. They announced that this model is a straight-up a <strong>reasoning model </strong>and that it would take more time to reason and answer your questions providing more accurate answers. They state that they have put more compute time into RL (Reinforcement Learning) than previous models to generate coherent <strong><em>chains-of-thoughts⁴</em></strong>. Essentially, they have trained the chain of thought generation process using Reinforcement learning (to generate and hone its own generated chain of thought process). In the o1 models, the engineers were able to ask the model questions as to why it was wrong (whenever it was wrong) in its chain-of-thought process and it could identify the mistakes and correct itself from them. The model could question itself and have to reflect (see “Reflection in LLMs”) on its outputs and correct itself.</p><p>In another video “<a href="https://youtu.be/3BkQI3nIiB8?feature=shared">Reasoning with OpenAI o1</a>”, <a href="https://www.linkedin.com/in/jerry-tworek-b5b9aa56/">Jerry Tworek</a> demonstrates how previous OpenAI and most other LLMs in the market tend to fail on the following prompt:</p><blockquote>“Assume the laws of physics on earth. A small strawberry is put into a normal cup and the cup is placed upside down on a table. Someone then takes the cup and puts it inside the microwave. Where is the strawberry now? Explain your reasoning step by step.”</blockquote><p>Legacy GPT-4 answers as follows:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*uavcpaSd8YTpwoLYcj6_YA.png" /><figcaption>Figure 1: GPT-4 getting the strawberry in a cup question wrong</figcaption></figure><p>The relatively newer GPT-4o also gets it wrong:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vMq3ATGYKkUP4Nehl0JBow.png" /><figcaption>Figure 2: GPT-4 o getting the strawberry in a cup question wrong</figcaption></figure><p>GPT o1 gets the answer right:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*tdyoYYjdtOW_Es9TT97E7w.png" /><figcaption>Figure 3: GPT o1 getting the answer to the strawberry in a cup question right</figcaption></figure><p>If you click on the dropdown at the beginning of the model’s response (see Figure 4), you know that it elicits its thought process (chain-of-thought), and the researchers at OpenAI claim that the o1 model has been trained with reinforcement learning to get this chain of thought better. Also, it’s interesting to note that <a href="https://www.linkedin.com/in/jason-wei-5a7323b0/">Jason Wei</a> (you can see him sitting third from the right on the bottom row in the video “<a href="https://youtu.be/3k89FMJhZ00?feature=shared">Building OpenAI o1</a>”), the author of the chain-of-thought paper⁴ that he published back at Google, is now an OpenAI employee that is working on the o1 model to integrate the chain-of-thought (that he discovered at Google) process into this model.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/715/1*6FNw5V_CasKHIEv-ybcaxg.png" /><figcaption>Figure 4: Chain-of-thought elicitation by GPT o1</figcaption></figure><p>Now, let’s get back to the counting question that my team found out about as part of my assignment.</p><blockquote>How many r’s are in the word strawberry?</blockquote><p>Let’s run this question on GPT-4o:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*TeDTDQ3vgBE_fOqTIOgHwQ.png" /><figcaption>Figure 5: GPT4o getting the number of r’s in strawberry question wrong.</figcaption></figure><p>A very simple counting problem that it get’s wrong.</p><p>Let’s run this on the new GPT o1:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*q7ObqVwDIHUY8hfoNZnHAQ.png" /><figcaption>Figure 6: GPT o1 getting the number of r’s in strawberry question right.</figcaption></figure><p>GPT o1 gets the answer right by thinking for a couple of seconds. The researchers at OpenAI say that it goes through its response repeatedly and thinks its way to the right answer. There does seem to be really significant improvements in terms of the model’s ability to solve a lot of academic exam questions.</p><p>Anyways after I opened up X.com (Formerly Twitter), I came across several people showcasing their attempts at trying to make the o1 model fail. This is a fascinating one I came across (this <a href="https://x.com/creeoer/status/1834588136749388035">tweet</a> by @creeor) where the model fails to answer a trivial question in which the answer lies in the question itself. So I tried the exact same prompt on my account and it gave me the wrong answer (see Figure 7).</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*BB8M867R7Xkcdi1s2Lc-TA.png" /><figcaption>Figure 7: OpenAI o1 is still getting simple riddles wrong when you tweak the riddle. Shows that the models still rely on a lot of the memorization that happened during it’s training and isn’t fully using it’s reasoning capabilities.</figcaption></figure><p>When I ask it what this classic riddle is that it is talking about it tells me about a riddle that it memorized from the internet. It’s interesting to see how these models can sometimes fall back on memorized content rather than truly reasoning through a problem. Despite the significant advancements and <a href="https://cdn.openai.com/o1-system-card.pdf">benchmark</a> improvements, there are still areas where AI models struggle, especially with tasks that require deeper reasoning or understanding context in a nuanced way. While benchmarks can show progress, real-world applications often reveal the limitations. It’s through continuous testing, feedback, and real-world use cases that these models can be further refined.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*UDOjGGL0ugZ7yMu5rpOfQg.png" /><figcaption>Figure 8: o1 blindly answers from a riddle it memorized. It doesn’t read the question that was given to it and try to answer it as presented.</figcaption></figure><p>There was a <a href="https://medium.com/@aliborji/a-categorical-archive-of-chatgpt-failures-2c888805d3c3">compilation</a> of ChatGPT failures done about a year and a half back. Compilations of model errors are invaluable for understanding and improving AI systems. I’m sure people will come up with another compilation of errors for the o1 model soon.</p><p>Although I agree entirely that the chain-of-thought process benefits both AI and human learning, real learning indeed comes from experience and making mistakes.</p><p>I will continue to keep posting my findings on the o1 model on my Medium page. Follow my account to keep posted. And thank you for taking the time to read my Medium post.</p><h3>References:</h3><p>[1] Arkoudas, Konstantine. “GPT-4 can’t reason.” <em>arXiv preprint arXiv:2308.03762</em> (2023).</p><p>[2] Aghzal, Mohamed, Erion Plaku, and Ziyu Yao. “Look Further Ahead: Testing the Limits of GPT-4 in Path Planning.” <em>arXiv preprint arXiv:2406.12000</em> (2024).</p><p>[3] Kambhampati, Subbarao, et al. “LLMs Can’t Plan, But Can Help Planning in LLM-Modulo Frameworks.” <em>arXiv preprint arXiv:2402.01817</em> (2024).</p><p>[4] Wei, Jason, et al. “Chain-of-thought prompting elicits reasoning in large language models.” <em>Advances in neural information processing systems</em> 35 (2022): 24824–24837.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=99396d641fff" width="1" height="1" alt=""><hr><p><a href="https://medium.com/data-science/openai-o1-the-enigmatic-force-that-will-reshape-every-knowledge-sector-that-we-know-of-or-99396d641fff">OpenAI o1: Is This the Enigmatic Force That Will Reshape Every Knowledge Sector We Know?</a> was originally published in <a href="https://medium.com/data-science">TDS Archive</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Probability for Machine Learning — My Journey re-learning Probability to understand ML Research]]></title>
            <link>https://medium.com/@abhinavp_41237/probability-for-machine-learning-my-journey-re-learning-probability-to-understand-ml-research-ab2b7eeda175?source=rss-5ffa75f9ef71------2</link>
            <guid isPermaLink="false">https://medium.com/p/ab2b7eeda175</guid>
            <category><![CDATA[probability]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[machine-learning]]></category>
            <dc:creator><![CDATA[Abhinav Prasad Yasaswi]]></dc:creator>
            <pubDate>Fri, 06 Sep 2024 05:14:51 GMT</pubDate>
            <atom:updated>2024-09-06T05:14:51.180Z</atom:updated>
            <content:encoded><![CDATA[<h3>Probability for Machine Learning — My Journey re-learning Probability to understand ML Research</h3><figure><img alt="Stock Image (generated by Gemini) of a paper consisting of mathematical calculations that involve probability." src="https://cdn-images-1.medium.com/max/1024/1*OWlmt3mdiB08immlE8L_Lw.jpeg" /></figure><h3>Motivation</h3><p>I am currently in the third semester of my Master&#39;s degree in Computer Science. It so happens that I’ve taken three ML courses (Frontier Topics in Generative AI, NLP, Trustworthy Human-ML Interaction), and most of these courses demand a solid understanding of mathematical concepts that we either learned back in high school or are a little more advanced than freshman college math.</p><p>I am embarking on a journey of writing articles on my Medium profile to relearn and revise concepts I’ve learned in the past. I hope to better understand the math behind a lot of ML frameworks and architectures to further be able to contribute in a more meaningful manner.</p><p>I hope to stay dedicated and write at least one article a day for the next year or so. I appreciate all the people who find my articles helpful and take the time to read them.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*bfEr3lIitsat7mH1vQaI7Q.png" /><figcaption>Figure 1: A Probability and Statistics Roadmap for NLP found on nlp-roadmap by graykode on GitHub.</figcaption></figure><p>I will be using the roadmap found on:</p><blockquote><a href="https://github.com/graykode/nlp-roadmap/tree/master">graykode/nlp-roadmap: ROADMAP(Mind Map) and KEYWORD for students those who have an interest in learning NLP (github.com)</a></blockquote><p>I will go in the order:</p><ol><li>Basic</li><li>Bayesian</li><li>Sampling</li><li>Information Theory</li><li>Model</li></ol><p>Let’s get on the Journey!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ab2b7eeda175" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>