Update README and challenge docs

Author: gbg141

README.md

@@ -33,22 +33,10 @@ Assess how your model compares against state-of-the-art topological neural netwo
   <a href="#mag-references">References</a> 
 </p>
 
-## 🏆 Announcing the Topological Deep Learning Challenge 2025!
 
-We are excited to announce that `TopoBench` is hosting the **Topological Deep Learning Challenge 2025**, as part of the Topology, Algebra, and Geometry in Data Science (TAG-DS) 2025 conference. This year's theme is **"Expanding the Data Landscape"**: 
-participants are invited to implement a new or existing dataset within TopoBench.
-<p align="center">
-  <img src="resources/2025_challenge_flyer.png" alt="TDL Challenge 2025 Flyer" width="800">
-</p>
-
-#### Why Participate?
-* 🌟 Co-author a white paper for the Proceedings of Machine Learning Research (PMLR).
-* 🌟 Win cash prizes up to $800 USD, sponsored by [Arlequin AI](https://www.arlq.ai/).
-* 🌟 Secure internship opportunities at UC Santa Barbara and EPFL.
-
-**Deadline for submission:** November 25th, 2025 (AoE) 
+---
 
-For more details, rules, and to get started, please visit the [`link to the challenge website`](https://geometric-intelligence.github.io/topobench/tdl-challenge/index.html).
+> 🏆 The **TAG-DS Topological Deep Learning Challenge 2025** has concluded! A huge shotout to all participants. Check out the winners and honorable mentions on [`the challenge website`](https://geometric-intelligence.github.io/topobench/tdl-challenge/index.html).
 
 ---
 

docs/tdl-challenge/index.rst

@@ -8,6 +8,29 @@ Welcome to the **Topological Deep Learning Challenge 2025: Expanding the Data La
 .. seealso::
    Link to the challenge repository: `<https://github.com/geometric-intelligence/TopoBench>`__.
 
+🏆 Winners
+----------
+
+**Category A.1: Broadening Benchmarks**
+
+* **Winner:** *Amiiza team*. Their implementation of *ATLAS Top Tagging* brings a novel benchmark from the particle physics community into the TDL ecosystem.
+* **Honorable Mentions:** *MappingComplexityLab* for the HypBench Dataset and *Loris* for the Graphland Benchmark.
+
+**Category A.2: Natively Higher-Order Datasets**
+
+* **Winner:** *Hugo Walter*. He implemented the *Cornell LabeledNodesDataset*, consisting of 8 single-labeled hypergraph datasets, representing a significant step forward for hypergraph benchmarking.
+* **Honorable Mentions:** *IgPA* for the Conjugated Molecule dataset and *TJPaik* for Analog Circuits implementations.
+
+**Category B.1: Large-Scale Inductive Infrastructure**
+
+* **Winner:** *DLLB team*. This team successfully tackled the challenge of migrating to ``OnDiskDataset`` support in the transductive setting.
+* **Honorable Mention:** *DLLB team*. The team received a special acknowledgement for their impressive strides in scalable data loading for the inductive setting as well.
+
+**Category B.2: Pioneering New Benchmark Tasks**
+
+* **Winner:** *DLLB team*. They introduced Link Prediction to the benchmark suite, establishing a new standard for the field.
+* **Honorable Mention:** *NeuroTriangles team*. Recognized for proposing higher-order cell tasks on a neuroscience dataset.
+
 
 Motivation
 ----------