University Research Collaboration

WHY AGRICULTURAL RESEARCH OFTEN FAILS TO REACH FARMERS — A Consultant’s Perspective Having worked with dozens of cooperatives, farmer groups, and agrifood projects across Kenya, I’ve seen a pattern that’s hard to ignore: Agricultural research is abundant. Impact on the ground? Minimal. Why? Research is often academic, not practical. Brilliant findings end up in journals, not in farmers’ hands. Most farmers I work with have never seen or heard of the latest research that could transform their yields or earnings. Top-down approaches dominate. Solutions are designed in labs or research stations with minimal farmer involvement. Yet, farmers are the experts of their own environments. Poor extension linkages. Even when good innovations exist, there’s a huge gap between research institutions and grassroots extension systems. As consultants, we often end up "translating" research that should have been made farmer-friendly from the start. No market lens. Research tends to focus on production. But farmers ask: “Will it sell? Is it profitable?” Without market integration, innovation is just theory. Feedback is ignored. Farmers are rarely involved in evaluating what works or doesn’t. We need more participatory learning, less top-down training. From a consultant’s view, the solution is not just more research—but more relevant, inclusive, and actionable research. Let’s invest in: Co-creating with farmers, Bridging research with market realities, Translating findings into practical guides, audio-visuals, and demos, Strengthening extension and private sector partnerships. The knowledge exists. The gap is in the approach. Farmers don’t need more data—they need results. #Agriculture #FarmersFirst #ResearchToImpact #KenyaFarming #AgriConsulting #FoodSystems #ValueAddition #DairyDevelopment #ExtensionServices #AgriPolicy #AfricanAgriculture

Academia says it values global knowledge. But try publishing from Nairobi instead of New York. Nature just introduced citation diversity statements. Authors should confirm they cited diverse researchers. These statements sound progressive but they miss the deeper truth: Academic publishing still sorts by passport, not merit. Aigner et al. shows 3 barriers blocking global researchers: 1. Gatekeeping by Location Developing country = 10x lower acceptance at top journals. Citation quality doesn't explain this gap. Same citation counts get different treatment Location predicts publication more than merit. Top-10 journals: 15.5% acceptance for U.S. authors But only 1.4% for developing country authors It’s not the paper quality. 2. Network Segregation Top-25 journal representation unchanged since 1980. All growth happens in journals ranked 100th or lower. Our profession prioritizes passport over content. Elite journals maintain closed networks. Geography over research quality. Your network determines your visibility. 3. Knowledge Exclusion Researchers from low-income countries publish less than their economic impact. Missing voices means missing insights. Questions from the Global South remain unasked. Local knowledge gets systematically excluded. Our research narrows when it should expand. We're losing vital perspectives. Recognize these barriers to navigate them: Strategic collaborations can open closed networks. Target journals outside the traditional hierarchy. Document bias when you encounter it. Until we fix these structural gaps: Citation diversity statements are just for show. Save this if you publish, peer review, or teach. These numbers matter. Nature’s post: https://lnkd.in/eNBDdHf6 Aigner paper: https://lnkd.in/emgib6ZJ Tips to get published & cited: https://lnkd.in/eVtvPfhu #research #publishing #phd

Is It Possible to Do Research Facing This Wall? In many parts of the world, doing science is not a clear academic path, but rather an obstacle course. These are some of the most common (and often normalized) barriers faced by researchers: 🧱 1. Disproportionate teaching load Classroom responsibilities consume what should be dedicated research time. And no, better time management is not the answer. 🧱 2. Excessive bureaucracy and absurd controls More time spent justifying your work than actually doing it. Administrative overload can drain more energy than the research itself. 🧱 3. Scarce and unstable funding Research lines depend on unpredictable calls, shifting criteria, or politicized processes. Without continuity, there's no sustainability. 🧱 4. Outdated or inaccessible equipment How can we innovate without access to proper labs, software, or updated data? 🧱 5. Lack of institutional recognition Sometimes publishing in top-tier journals matters less than organizing events or pleasing internal politics. Excellence gets sidelined. 🧱 6. Weak engagement with society Without meaningful university-industry-community collaboration, research becomes isolated. The myth that research is "just theory" gets reinforced. 🧱 7. Job insecurity Many leave academia altogether. Others stay but burn out, becoming passive gatekeepers instead of mentors for new generations. 📣 How many of these barriers have you faced? How many go unnoticed by decision-makers? Science needs more than talent. It needs real conditions. And in too many places, we’re still far from providing them. #AcademicResearch #FairEvaluation #SciencePolicy #HigherEducation #SupportResearchers #ScienceMatters

Very promising! A new open-source platform for research on Human-AI teaming from Duke University uses real-time human physiological and behavioral data such as eye gaze, EEG, ECG, across a wide range of test situations to identify how to improve Human-AI collaboration. Selected insights from the CREW project paper (link in comments): 💡 Comprehensive Design for Collaborative Research. CREW is built to unify multidisciplinary research across machine learning, neuroscience, and cognitive science by offering extensible environments, multimodal feedback, and seamless human-agent interactions. Its modular design allows researchers to quickly modify tasks, integrate diverse AI algorithms, and analyze human behavior through physiological data. 🔄 Real-Time Interaction for Dynamic Decision-Making. CREW’s real-time feedback channels enables researchers to study dynamic decision-making and adaptive AI responses. Unlike traditional offline feedback systems, CREW supports continuous and instantaneous human guidance, crucial for simulating real-world scenarios, and making it easier to study how AI can best align with human intentions in rapidly changing environments. 📊 Benchmarking Across Tasks and Populations. CREW enables large-scale benchmarking of human-guided reinforcement learning (RL) algorithms. By conducting 50 parallel experiments across multiple tasks, researchers could test the scalability of state-of-the-art frameworks like Deep TAMER. This ability to scale the study of the interaction of human cognitive traits with AI training outcomes is a first. 🌟 Cognitive Traits Driving AI Success. The study highlighted key human cognitive traits—spatial reasoning, reflexes, and predictive abilities—as critical factors in enhancing AI performance. Overall, individuals with superior cognitive test scores consistently trained better-performing agents, underscoring the value of understanding and leveraging human strengths in collaborative AI development. Given that Humans + AI should be at the heart of progress, this platform promises to be a massive enabler of better Human-AI collaboration. In particular, it can help in designing human-AI interfaces that apply specific human cognitive capabilities to improve AI learning and adaptability. Love it!

WHO's guidance provides a framework for developing equitable clinical research capacity around the world. https://lnkd.in/eqKYBWkh As we developed it, the consultation process highlighted the importance of considering an equity and justice perspective, with a particular focus on under-represented populations and countries with a high burden of illness. It has long been known that authors from LMIC settings tend to be under-represented for first or last author positions in authorship of clinical trials where there are collaborations with high income country authors. South-south collaborations tend to give greater first and last authorship representation to those working in the countries of greatest focus of the trials. https://lnkd.in/eMVWSVTA I see a new article in press confirming this finding in specific cancer domains https://lnkd.in/egVahFap These papers also identify a number of trials where there are no authors at all from the LMIC countries of focus. This is very far from the equitable research capacity development we are calling for. Do those of you based in LMIC see improvements in this area of representation? What are the next steps in addressing this problem?

🤖 𝐓𝐡𝐞 𝐅𝐮𝐭𝐮𝐫𝐞 𝐢𝐬 𝐌𝐮𝐥𝐭𝐢-𝐀𝐠𝐞𝐧𝐭: 𝐅𝐫𝐨𝐦 𝐒𝐢𝐧𝐠𝐥𝐞 𝐑𝐨𝐛𝐨𝐭𝐬 𝐭𝐨 𝐂𝐨𝐥𝐥𝐚𝐛𝐨𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈 𝐒𝐲𝐬𝐭𝐞𝐦𝐬 Just reviewed a fascinating survey paper on "Multi-Agent Embodied AI: Advances and Future Directions". 𝐊𝐞𝐲 𝐈𝐧𝐬𝐢𝐠𝐡𝐭𝐬:- 🔹 𝐅𝐫𝐨𝐦 𝐒𝐨𝐥𝐨 𝐭𝐨 𝐒𝐲𝐦𝐩𝐡𝐨𝐧𝐲- While most research has focused on single-agent systems, real-world applications demand multiple agents working together. Think warehouse robots, autonomous vehicle fleets, and healthcare teams. 🔹 𝐓𝐡𝐞 𝐂𝐡𝐚𝐥𝐥𝐞𝐧𝐠𝐞 𝐢𝐬 𝐑𝐞𝐚𝐥- Multi-agent systems face unique hurdles:- - Asynchronous decision-making (agents operating at different speeds). - Heterogeneous capabilities (drones + robotic arms + vehicles). - Dynamic environments where team composition constantly changes. 🔹 𝐆𝐚𝐦𝐞-𝐂𝐡𝐚𝐧𝐠𝐢𝐧𝐠 𝐓𝐞𝐜𝐡𝐧𝐨𝐥𝐨𝐠𝐢𝐞𝐬- - Integration of Large Language Models for natural language coordination. - Generative models for better task planning and allocation. - Advances in MARL (Multi-Agent Reinforcement Learning). 🔹 𝐏𝐫𝐚𝐜𝐭𝐢𝐜𝐚𝐥 𝐀𝐩𝐩𝐥𝐢𝐜𝐚𝐭𝐢𝐨𝐧𝐬- - Smart manufacturing with collaborative robot teams. - Autonomous driving with vehicle-to-vehicle coordination. - Healthcare systems with AI assistants working alongside medical staff. The paper excellently bridges the gap between theoretical advances and real-world implementation. It's clear that the future belongs to systems where AI agents don't just operate independently but truly collaborate. 𝐖𝐡𝐚𝐭 𝐞𝐱𝐜𝐢𝐭𝐞𝐬 𝐦𝐞 𝐦𝐨𝐬𝐭? The emphasis on human-AI collaboration and the development of specialized benchmarks for testing multi-agent scenarios. #MultiAgentAI #EmbodiedAI #MachineLearning #Robotics #Innovation #FutureTech

New article in #JDS: What’s the right dose of rumen-protected choline for dairy cows? Happy to share our latest publication—produced in collaboration with Dr. Barry Bradford and international team—where we performed a systematic review and dose-response meta-analysis of #rumen_protected_choline supplementation in dairy cows. 🔹 Key finding: Supplementing 13–14 g/d RPC optimizes milk yield (+1.3 kg/d) and intake, while higher doses can boost fat-corrected milk yield even further. This work provides practical guidance for dairy nutritionists worldwide. Grateful for the opportunity to work with Dr. A. H. Piray, Dr. Rezaei-Ahvanooei, Dr. Ma, and Dr. Bahrampour on this impactful project! Read more: https://lnkd.in/ebHnKmrv #dairy #choline #metaAnalysis #dairycows #dairyscience

Check out the latest from MIT EQuS and Lincoln Lab published in Nature Physics! In this work, we use a 4x4 array of superconducting transmon qubits to emulate the dynamics of charged particles moving through electromagnetic fields. https://lnkd.in/eC5mANRH https://rdcu.be/dYAVC Superconducting qubit arrays natively emulate the Bose-Hubbard model in the absence of a magnetic field. In this work, we develop a scheme to parametrically couple adjacent qubits such that their coupling reflects an adjustable synthetic magnetic vector potential. We verify that spatially varying the vector potential then creates a synthetic magnetic field via Gauss’s law for magnetism, and varying the vector potential in time creates an electric field via Faraday’s law of induction. Our work enables superconducting qubit arrays to simulate a wide range of condensed matter physics such as the Hall effect. Congratulations Ilan Rosen, Sarah Muschinske, and all co-authors with the Massachusetts Institute of Technology, MIT EQuS Group, MIT Lincoln Laboratory. #quantumcomputing. MIT Center for Quantum Engineering, MIT School of Science, MIT School of Engineering, MIT Department of Physics, MIT EECS, Research Laboratory of Electronics at MIT, MIT xPRO, #quantumcomputing. 

Seeking overseas Admission/ Collaboration After publications of n number of research articles in indexed journals, researchers from developing countries / least developing countries are seeking exposure in an established laboratory, preferably in a developed country. Most of them end up with a situation that they don't get response back on possible placement / collaboration from those laboratories. There is a need to revisit the issue. These labs get large number of requests, and it is not possible to accommodate all of them. Normally, researchers talks about their strength and seek placement/ collaboration. Few also expect remuneration/ fellowship and this may or may not motivate established laboratory to respond. So, to improve chances of a new proposal to be considered, it would be useful if one study the kind of work being done in that established laboratory and also introduce her/his research strength in first go along with a suggestion how it will further augment work of established laboratory and also suggest them how it could be financed (by way of existing programs/ schemes, which requires a little work from the host laboratory and it is external funding to them. This may attract better response

Low emissions milk can be produced without sacrificing on farm profitability. This was a key takeaway from an analysis of real farm data released by DairyNZ.   The study was a collaboration between DairyNZ, Fonterra and LIC, which connected together large industry datasets, with data from around 8,000 farms. It allowed detailed analysis of the relationship between emissions and profitability, with the results from the Waikato and Canterbury regions published.   A summary of the findings were: 🐄 There was no significant relationship between profit and emissions intensity which essentially means high profit farms can either be low or high emissions intensity (which is mainly driven by farm management decisions).   🐄 The key to a high profit and low emission intensity farm, is using low footprint feed to achieve good, but not exceptional, milk production per kg liveweight.   🐄 Low footprint feed is homegrown, uses nitrogen efficiently, and supplements with lower embedded emissions.   🐄 Aiming for low emissions intensity through increased production per cow without focusing on the footprint of the feed, is likely to have undesirable consequences on other key outputs e.g., profitability, total emissions.   🐄 High profit farms with low emissions intensity can be found anywhere within each region.   🐄 All farms have opportunities to lower emission intensity, without compromising other outputs. More research using the datasets is planned which will help inform farmers of the practical areas they can focus on to help improve on farm emissions intensity and also profitability.