Update README.md

README.md

@@ -20,22 +20,6 @@ short_description: Reshaping businesses for the agentic era.
     </div>
   </header>
 
-  <p><strong>Foaster.ai</strong> is a French start-up focused on reshaping businesses for the agentic era. At <em>Foaster Labs</em>, our Werewolf Benchmark studies how LLMs behave under social pressure: leadership, bluffing, and resistance to manipulation.</p>
-
-  <div style="display:flex;align-items:center;gap:10px;margin:14px 0 22px;">
-    <a href="https://huggingface.co/spaces/Foaster/Werewolf_benchmark"
-       style="padding:10px 14px;border:1px solid #e5e7eb;border-radius:10px;text-decoration:none;">
-      🔗 Full leaderboard →
-    </a>
-  </div>
-
-  <h3 style="margin:0 0 8px;">Results — Podium (role-conditioned Elo)</h3>
-  <p style="margin:0 0 10px;color:#64748b">ELO-W = wolf (manipulation power) · ELO-V = villager (manipulation resistance)</p>
-
-  <ul style="margin:0 0 8px 18px;">
-    <li>🥇 <strong>GPT-5 (OpenAI)</strong> — ELO 1492 (W 1508 · V 1476), win rate 96.7%, 60 matches</li>
-    <li>🥈 <strong>Gemini 2.5 Pro (Google)</strong> — ELO 1261 (W 1163 · V 1360), win rate 63.3%, 60 matches</li>
-    <li>🥉 <strong>Gemini 2.5 Flash (Google)</strong> — ELO 1188 (W 1103 · V 1273), win rate 51.7%, 60 matches</li>
-  </ul>
+  <p><strong>Foaster.ai</strong> is a French start-up focused on reshaping businesses for the agentic era. At <em>Foaster Labs</em>, we study LLMs and their personalities through original benchmarks and exploration frameworks.</p>
 
 </div>