Datasets:
OSC open discovery: CHf2 (1CHf2-1, OSC-01938) new material outside active set
Thanks for the Discovery proposal — new candidates outside the active set are welcome. We will downfold CHf2 (OSC-01938) with the public pipeline/ (SCF → downfolding → solver) to set its official value and report back.
Update: we downfolded CHf2 (OSC-01938) on our own DFT — canonical N(E_F) ≈ 1.89, filling δ ≈ +0.278 → score 33.2, currently the top entry. Honest caveat: two bands cross E_F in CHf2, so the score is sensitive to the active-band choice; we are verifying band-selection robustness (multi-band / Wannier) before finalizing. Thank you for a strong Discovery proposal.
Thank you for the quick downfold and for the transparent note on the two-band sensitivity — we look forward to the multi-band / Wannier robustness check.
One question on attribution. CHf2 was proposed in this PR (opened 2026-10-01 01:35 UTC) while it was outside the active set. After it was added to the active set, the same material can be submitted by anyone through the Space, and it now appears on the leaderboard under several names with an identical verified score.
Could you clarify how ties on the same material are resolved for the Grand and Best Discovery prizes — is the original proposer (earliest PR / timestamp) the credited entry?
Thanks again for running an open challenge.
Discovery credit for CHf2 is yours as the first proposer. Once a material enters the active set, later identical submissions receive the canonical score but not the Discovery credit; Grand/Discovery go to the first verified proposer, and ties on an identical canonical score are broken by submission time. (CHf2 itself is still under multi-band robustness review before its value is finalized.)