explore | Grilled Cheese

In SOAP note evaluation, lexical methods estimated 35% hallucination, but inference-aware evaluation reduced this to 9%, separating valid clinical reasoning from true errors. #SOAPNotes #ClinicalNLP #Evaluation doi.org/10.48550/arx...

📊 Impact #evaluation informs action in Mozambique 🇲🇿.
📉 6.7% reduction in severe stunting, alongside better sanitation, diets & child health, was a result of converging programming in Sofala province, evaluation shows. Findings are already shaping a new nutrition programme.
🔗 bit.ly/4yfYOvv 👈
#UNICEF

On #AlliedHealthProfessionalsDay @penninecarenhs.bsky.social Knowledge Service celebrating the #research #innovation & #evaluation of all our #AHPs supporting through the Knowledge & Library Hub tinyurl.com/3z8ntfh3 @laurajeffreys.bsky.social @sfstringwell.bsky.social @jackiekindell.bsky.social

Great to see researchers pushing the boundaries involving older people in #research...

... and evaluating it!

Engaging With a #ParticipatoryResearch Team: #Evaluation of Involving Older Co‐Researchers With Experience of Falls in the SeFallED Project onlinelibrary.wiley.com/doi/full/10....

Rethinking the CES Board Structure: Join a Member Town Hall

CES is proposing a governance structure built for today, and we want to hear from you.
buff.ly/TOph2n9
#CES #evaluation

Refonder la structure du CA de la SCÉ: une réunion d'information pour les membres

La SCÉ propose une structure de gouvernance adaptée aux réalités d'aujourd'hui, et nous souhaitons connaître votre avis.
buff.ly/MtV60FP
#SCE #evaluation

Some tips for building a panel of agentic AI judges

agenticaiinsights.substack.com/p/deeper-div...

#agenticAI #ai #evaluation #llms

📝 New blog post: Twenty-Two Agents, and the Lead Is Gone

Two days ago Sonnet 5.5 had won every blind pass of my agent evaluation. Then Haiku 5.5 came out, I found $18 of forgotten OpenR...

https://blog.arda.tr/blog/2026-10-12-twenty-two-agents-and-the-lead-is-gone

#ai #agents #evaluation #dev

📝 Survey: How is demand for evaluation changing globally?

Participate in this 10-minute survey & help identify emerging trends, priorities & opportunities for the international #evaluation community. Responses will be anonymous.

🗓️ Deadline: Oct. 15, 2026
👉 Take the survey: bit.ly/3UlGgMs

#airport #arrested #assault #attack #charge #evaluation #hospital #Jamaica #John F. Kennedy #Lin #Los Angeles #New Jersey #New York #Big Apple 🗽 #Police
🌐 www.newsmart.ch

📢Meaningful consent is at the heart of ethical evidence activities involving children.
Join #UNICEF & international experts to discuss consent, assent, agency, evolving capacities, practical challenges & good practices.
📅20 Oct | 14:30 CET
🔗Register: bit.ly/consentinres...
#Evaluation #Research

What does it take to turn promising ideas into solutions that can achieve sustainable impact at scale?
Watch this new video for an overview of the #Evaluation of Innovation in #UNICEF and its key findings and recommendations

🌍 On #DayOfTheGirl, #evaluations from Benin & Nigeria show that girls succeed when quality education is paired with mentoring, women's leadership & community support. Evidence for action: 🔗 bit.ly/4jEdGA6 | bit.ly/3TFQwyI 👈 #IGD2026 #GirlsEducation #Evaluation #EvidenceForAction

#DayOfTheGirl: A #UNICEF #evaluation evidence shows promising multisectoral models & youth engagement supporting adolescent girls leadership & empowerment in East Asia Pacific. Gaps remain in prioritization, systematic tracking, & disaggregated data. Girl-centred action is essential.🔗 bit.ly/4uAg7XA

🤖 AI can help us reach the unreached, but can we do so without causing harm? At #FL26, UNICEF Evaluation’s Zlata Bruckauf joined experts to explore responsible AI in fragile settings and what it means for the future of evaluation. #AI #Evaluation #UNICEF
🔗 fl26.exordo.com

Most health LLM studies tested exam-style QA, not real care: only 5% used real patient data, and 95.4% focused on accuracy. #LLM #HealthcareAI #Evaluation doi.org/10.1001/jama...