Update dataset card metadata, link paper and GitHub

Hi! I'm Niels from the Hugging Face community science team.

This PR improves the dataset card by:
- Adding `task_categories` and `license` metadata.
- Adding tags for better discoverability (`biology`, `antibody-design`).
- Linking the dataset to its official paper page on the Hugging Face Hub.
- Adding a link to the official GitHub repository.
- Maintaining the detailed dataset structure and evaluation metrics documentation.
- Including a sample usage code snippet.

README.md

@@ -1,11 +1,25 @@
+---
+task_categories:
+- other
+license: cc-by-4.0
+tags:
+- biology
+- antibody-design
+- protein-structure
+- epitope-specific
+---
+
 # CHIMERA-Bench v1.0
 
 A unified benchmark for epitope-specific antibody CDR sequence-structure co-design.
 
-**Paper**: CHIMERA-Bench: A Benchmark Dataset for Epitope-Specific Antibody Design (ICLR GEM Workshop 2026)
+**Paper**: [CHIMERA-Bench: A Benchmark Dataset for Epitope-Specific Antibody Design](https://huggingface.co/papers/2603.13431) (ICLR GEM Workshop 2026)  
+**Code**: [GitHub - mansoor181/chimera-bench](https://github.com/mansoor181/chimera-bench)
 
 ## Dataset Summary
 
+CHIMERA-Bench (CDR Modeling with Epitope-guided Redesign) is a curated, deduplicated dataset of **2,922** antibody-antigen complexes with epitope and paratope annotations. It provides a standardized evaluation protocol for antibody design tasks.
+
 | Property | Value |
 |----------|-------|
 | Complexes | 2,922 |
@@ -39,7 +53,7 @@ chimera-bench-v1.0/
 
 ## Complex Features Format
 
-Each `.pt` file is a Python dict with:
+Each `.pt` file is a Python dict containing:
 
 **Sequences**
 - `complex_id`: str -- unique identifier ({pdb}_{Hchain}_{Lchain}_{Agchain})
@@ -57,35 +71,19 @@ Each `.pt` file is a Python dict with:
 - `contact_pairs`: list of (ab_chain, ab_resid, ab_resname, ag_chain, ag_resid, ag_resname, distance)
 
 **Numbering**
-- `numbering`: dict with `imgt` and `chothia` sub-dicts, each containing `heavy` and `light` lists of (resnum, icode, aa) tuples
-- `cdr_masks`: dict with `imgt` and `chothia` sub-dicts, each containing `heavy` and `light` int lists (-1=framework; heavy: 0=H1, 1=H2, 2=H3; light: 3=L1, 4=L2, 5=L3)
+- `numbering`: dict with `imgt` and `chothia` sub-dicts
+- `cdr_masks`: dict with `imgt` and `chothia` sub-dicts (-1=framework; heavy: 0=H1, 1=H2, 2=H3; light: 3=L1, 4=L2, 5=L3)
 
 **Surface Features**
-- `ag_surface_points`: float32 (128, 3) -- sampled antigen surface points
-- `ag_surface_normals`: float32 (128, 3)
-- `ag_surface_curvatures`: float32 (128, 2) -- mean and Gaussian curvature
-- `ag_surface_chemical_feats`: float32 (128, 6) -- hydropathy, charge, H-bond donor/acceptor, aromaticity, polarity
-- Same for `heavy_surface_*` and `light_surface_*`
+- `ag_surface_points`, `ag_surface_normals`, `ag_surface_curvatures`, `ag_surface_chemical_feats`.
 
 ## Splits
 
-- **epitope_group**: clusters by epitope residue fingerprint; test set has epitope patterns unseen during training
-- **antigen_fold**: clusters by antigen identity; test set has entirely unseen antigens
-- **temporal**: splits by PDB deposition date; simulates prospective deployment
-
-Each split JSON has keys `train`, `val`, `test` mapping to lists of complex_id strings.
+- **epitope_group**: clusters by epitope residue fingerprint; test set has epitope patterns unseen during training.
+- **antigen_fold**: clusters by antigen identity; test set has entirely unseen antigens.
+- **temporal**: splits by PDB deposition date; simulates prospective deployment.
 
-## Evaluation Metrics
-
-| Group | Metrics |
-|-------|---------|
-| Sequence quality | AAR, CAAR, PPL |
-| Structural accuracy | RMSD (Kabsch-aligned CA), TM-score |
-| Binding interface | Fnat, iRMSD, DockQ |
-| Epitope specificity | EpiF1 (precision, recall, F1) |
-| Designability | n_liabilities (NG, DG, DS, DD, NS, NT, M motifs) |
-
-## Quick Start
+## Sample Usage
 
 ```python
 import torch, json, pandas as pd
@@ -105,6 +103,16 @@ print(f"Epitope residues: {len(feat['epitope_residues'])}")
 print(f"CDR-H3 (IMGT): positions where cdr_masks['imgt']['heavy'] == 2")
 ```
 
+## Evaluation Metrics
+
+| Group | Metrics |
+|-------|---------|
+| Sequence quality | AAR, CAAR, PPL |
+| Structural accuracy | RMSD (Kabsch-aligned CA), TM-score |
+| Binding interface | Fnat, iRMSD, DockQ |
+| Epitope specificity | EpiF1 (precision, recall, F1) |
+| Designability | n_liabilities (NG, DG, DS, DD, NS, NT, M motifs) |
+
 ## Citation
 
 ```bibtex
@@ -120,4 +128,5 @@ url={https://openreview.net/forum?id=PyZvVIJbSy}
 
 ## License
 
-Data: CC-BY 4.0. Code: MIT.
+- **Data**: CC-BY 4.0
+- **Code**: MIT