Upload 12 files

content/my_index/0.codes.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e05a65f022ed226a9587d68e3d4a78223063827a6ebeacd378f85ccd4bd6812
+size 2584860

content/my_index/0.metadata.json

@@ -0,0 +1,6 @@
+{
+    "passage_offset": 0,
+    "num_passages": 3485,
+    "num_embeddings": 645932,
+    "embedding_offset": 0
+}

content/my_index/0.residuals.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7dd71f9174a4bb85b47d58e5d154bea4d863e01db21d1fd5cc29bd58d72b81c
+size 82680496

content/my_index/avg_residual.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad239a2fb858b2f3a4f442295741d56e8c459eb86fe4d8ee48b1bd33b00ea8f6
+size 1205

content/my_index/buckets.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4e96a6f5e5966b22d9ad758f82f711ccb7d1cfc3145971885b35277875c9a0f
+size 2904

content/my_index/centroids.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d8b59a0db2b6955cf1d963346007e037e0e809f647cb3955073c35ca8f1efd0
+size 2098342

content/my_index/doclens.0.json

@@ -0,0 +1 @@
+[232,79,188,237,78,211,210,239,199,180,214,100,149,224,140,237,145,218,208,230,213,233,115,191,217,239,152,206,114,206,44,187,144,230,231,233,205,161,186,227,189,178,220,215,189,180,231,162,199,197,193,232,223,54,235,184,159,182,231,74,165,233,216,181,215,100,230,206,227,59,231,153,152,226,223,236,199,200,179,211,193,200,176,233,66,203,126,231,224,77,229,118,188,236,238,98,219,203,226,191,213,229,114,230,217,196,213,181,222,99,202,211,71,221,90,226,90,206,239,237,75,220,67,230,163,213,176,165,231,79,222,201,233,183,199,232,224,212,88,208,194,208,230,171,199,125,240,144,214,169,199,238,208,87,226,146,189,230,172,226,220,134,227,229,225,98,180,189,184,213,119,212,170,189,211,144,228,235,122,223,241,205,235,230,91,156,213,107,197,180,232,222,147,200,220,189,210,190,196,227,111,221,74,207,178,155,200,224,127,234,228,66,200,233,226,214,218,229,88,232,79,233,152,237,200,133,159,209,228,202,221,71,152,207,223,221,222,86,227,45,227,73,218,230,200,141,160,231,73,236,201,145,228,102,218,176,205,224,237,180,197,212,104,165,109,196,216,155,232,234,218,225,232,221,234,133,220,235,213,171,232,99,235,213,175,232,227,227,199,95,206,139,201,226,204,185,213,239,218,205,120,206,175,233,202,176,226,237,106,189,134,214,83,206,208,125,181,235,229,208,220,214,206,221,231,78,211,234,212,170,223,232,195,223,210,120,216,166,159,242,239,61,225,157,189,180,240,78,183,160,213,115,227,214,221,233,232,236,229,77,208,227,234,193,178,210,178,235,132,229,211,133,229,229,202,235,213,80,150,220,171,222,192,227,231,168,232,120,236,177,232,118,138,221,216,197,220,227,182,190,96,197,211,111,157,232,190,231,178,225,238,235,77,217,214,69,206,112,234,46,222,229,119,209,208,224,52,210,193,132,201,214,86,228,87,229,30,234,90,223,201,178,234,84,235,127,218,229,109,178,235,144,237,93,224,84,225,99,192,222,192,224,176,225,226,104,137,223,225,92,217,168,224,86,235,73,219,201,196,233,229,86,220,161,229,137,215,85,222,220,234,78,207,141,225,92,198,207,223,209,228,202,194,229,158,221,89,139,231,164,245,156,195,224,217,223,208,207,215,226,183,240,233,216,76,191,209,154,204,233,182,238,223,222,221,237,224,94,172,220,119,226,94,225,202,187,198,203,79,227,19,195,224,214,202,228,116,232,220,119,235,104,232,87,201,215,183,224,68,215,82,176,154,220,180,219,216,237,119,210,87,233,201,121,169,143,201,219,188,199,212,228,241,74,240,203,184,232,235,65,217,93,204,235,234,125,199,224,51,218,233,81,234,57,230,51,169,219,76,239,217,218,104,223,203,174,179,199,234,64,157,230,73,134,218,102,219,192,225,64,223,163,191,212,228,176,109,233,189,228,234,230,210,198,223,219,102,238,190,199,203,199,224,212,183,213,78,218,53,222,106,206,229,87,225,111,204,231,236,46,232,235,222,220,41,238,171,167,225,217,175,208,159,227,212,221,85,197,219,98,214,229,237,198,224,84,232,223,221,239,76,211,211,195,217,90,234,128,210,82,213,204,88,235,80,176,193,231,161,207,199,112,224,242,83,226,182,209,218,188,226,215,223,230,124,214,218,172,212,132,229,230,202,236,238,89,206,222,212,223,111,207,124,201,177,220,235,145,204,226,209,224,147,187,183,232,102,190,224,221,69,232,189,197,201,199,103,206,218,199,193,210,210,155,160,225,73,191,198,236,93,179,233,204,237,232,211,95,218,101,174,202,99,229,168,210,74,229,116,217,86,231,51,235,199,151,225,203,169,217,106,237,64,201,224,146,196,95,186,228,214,107,220,205,105,226,85,221,87,237,210,236,240,217,167,223,209,178,192,176,218,70,176,236,229,215,137,220,220,228,104,158,214,231,106,216,109,186,226,206,244,82,230,234,123,186,223,74,214,237,231,78,234,173,206,232,108,199,196,116,206,229,208,220,122,212,95,226,233,101,222,229,94,215,221,216,171,125,215,73,160,130,128,223,64,196,214,235,87,165,212,190,215,83,239,236,210,232,211,215,192,234,82,208,133,230,227,178,165,218,230,235,73,234,121,231,163,236,196,135,233,234,209,228,176,237,210,131,228,138,219,90,188,221,211,235,216,230,214,120,229,207,221,235,232,100,239,199,173,123,159,239,187,218,204,232,190,184,117,233,58,181,210,166,227,202,110,213,82,218,222,79,221,119,178,118,235,124,133,228,195,92,216,208,78,224,115,202,124,216,177,223,213,199,156,197,237,72,146,191,172,233,179,181,216,210,59,216,108,240,229,232,164,215,232,143,236,227,84,189,154,230,225,218,229,229,237,93,232,215,194,138,177,228,173,220,105,143,182,233,76,231,228,213,61,234,206,214,232,88,231,225,123,193,173,236,129,231,211,176,222,229,228,225,191,223,241,195,213,215,229,139,172,214,177,188,184,227,231,227,62,231,207,235,125,229,217,98,212,227,100,236,72,204,200,209,235,153,230,153,206,178,231,198,224,124,164,242,189,208,220,121,225,234,84,218,232,75,135,173,214,97,215,127,199,175,233,82,206,233,227,213,236,225,221,196,159,214,94,223,124,211,214,78,239,231,144,235,85,223,235,220,231,56,209,241,228,224,51,221,56,206,196,96,150,192,84,213,239,211,179,205,184,230,226,19,192,126,205,228,56,227,114,188,229,221,218,224,188,227,62,232,54,174,198,239,90,210,124,223,116,215,228,234,166,213,139,211,186,228,77,220,142,202,231,179,121,219,92,232,202,89,198,184,236,195,151,221,195,198,240,231,213,166,132,238,187,237,227,220,221,235,228,198,238,97,200,234,235,114,231,243,201,87,231,86,225,204,175,231,108,216,71,166,230,205,206,205,235,230,103,233,234,61,220,231,134,227,110,236,219,236,97,193,222,124,234,61,227,114,236,119,222,213,127,225,83,207,227,143,199,184,53,207,237,226,83,218,93,180,210,206,160,225,85,205,236,212,230,218,229,190,204,221,204,218,201,201,111,232,220,221,228,94,218,243,88,225,217,230,222,208,78,203,206,232,208,195,162,222,183,170,139,224,61,225,66,233,213,206,241,193,217,188,225,97,239,233,233,131,157,214,146,225,75,215,230,228,223,68,204,134,183,208,225,222,125,177,219,77,186,226,56,93,212,107,207,112,211,63,200,156,210,207,221,174,226,217,75,231,223,234,236,122,185,197,184,196,199,219,119,209,157,162,181,239,69,219,229,240,112,211,213,226,61,236,127,176,228,233,133,227,172,230,82,225,116,238,150,227,210,235,63,217,234,203,87,236,174,213,236,119,194,222,221,60,152,236,218,156,204,136,231,94,209,74,230,192,106,221,219,197,201,219,183,222,227,210,214,230,220,104,154,222,222,202,103,145,190,197,234,109,228,207,214,85,237,79,225,227,94,211,231,79,184,215,87,234,232,230,229,95,188,211,231,234,96,238,146,217,185,237,217,56,162,187,210,229,175,227,235,128,232,235,229,199,219,209,195,237,228,109,224,238,221,147,215,66,185,219,224,62,151,187,164,220,93,210,108,224,92,201,191,210,236,230,222,174,167,228,204,131,234,239,181,116,207,131,167,215,198,218,124,164,214,80,231,233,80,232,228,77,210,145,239,206,180,234,225,227,122,214,226,97,194,191,161,229,67,200,98,191,103,223,79,215,107,208,226,228,156,211,234,225,89,192,234,189,236,98,231,216,107,184,225,217,214,86,233,82,156,243,72,197,245,129,222,231,73,208,233,103,204,226,206,165,175,230,220,113,190,215,70,229,206,168,224,236,198,226,219,227,231,204,234,100,230,230,88,199,154,225,110,233,231,221,96,212,118,183,182,201,233,193,213,199,203,206,225,147,229,238,106,226,173,238,233,230,235,225,213,139,212,231,223,102,231,213,224,67,234,78,157,234,193,211,218,222,125,198,147,201,156,181,210,230,225,225,200,186,200,218,215,216,238,145,229,225,162,222,94,239,99,219,110,164,169,227,203,182,157,239,233,244,55,227,80,166,222,231,186,229,192,231,158,182,199,136,236,90,207,165,227,212,198,227,63,228,165,235,241,206,220,105,231,167,232,208,167,234,91,224,204,238,104,223,191,202,166,194,235,153,223,215,203,176,224,80,186,211,227,68,184,198,129,234,195,216,176,203,183,206,208,227,208,229,188,212,233,83,232,128,237,200,225,89,208,122,226,77,200,96,209,185,228,234,156,163,201,130,209,194,181,135,214,123,211,240,93,222,105,227,179,234,81,240,57,226,108,232,142,238,235,217,114,223,117,190,215,226,178,208,92,143,202,234,209,211,234,60,217,103,219,216,110,220,201,92,213,233,202,211,131,217,134,208,210,226,233,187,209,63,125,202,236,228,69,141,217,196,118,188,212,123,201,212,159,232,224,95,229,73,172,221,172,224,232,94,219,88,236,204,204,110,236,233,239,219,231,99,214,186,226,222,101,237,84,211,179,227,108,232,200,122,199,237,74,222,82,227,226,230,116,234,234,66,238,225,131,223,178,189,226,150,216,196,226,84,228,210,102,213,232,204,226,91,237,214,231,227,221,199,171,233,183,183,227,235,90,224,218,19,230,184,181,229,231,203,229,226,102,222,118,179,213,149,186,202,161,118,211,103,236,226,227,114,151,165,204,140,191,220,230,179,227,197,226,201,137,212,214,134,146,150,235,229,182,223,231,231,195,214,184,238,111,196,203,229,106,209,206,157,234,225,219,63,243,147,187,208,187,243,108,157,208,226,223,75,227,203,169,203,209,177,150,213,215,157,209,105,230,126,235,223,220,103,206,227,175,135,233,176,224,106,231,171,233,106,178,182,198,231,215,110,233,202,209,222,200,227,225,230,199,230,80,158,232,221,200,205,189,219,242,90,218,82,235,215,225,105,224,226,61,228,73,202,189,215,222,170,206,107,138,230,231,191,103,196,211,239,206,231,62,188,129,223,218,123,213,104,225,98,178,176,197,231,193,223,223,217,211,95,171,229,112,180,232,163,203,218,191,206,133,205,230,203,227,169,224,103,143,236,234,138,176,169,234,227,71,199,222,89,199,232,127,223,220,219,75,167,209,229,232,208,92,194,168,233,198,236,227,202,197,231,191,186,225,200,112,221,239,89,235,207,115,225,76,192,201,164,227,74,238,140,231,220,140,111,230,81,181,207,75,231,77,163,220,199,156,187,174,153,202,238,62,196,232,83,213,76,207,243,105,229,110,236,170,230,216,229,215,233,60,118,218,119,222,103,215,209,242,88,235,97,234,86,203,156,177,231,64,181,228,201,226,192,232,217,128,177,199,228,199,160,228,215,228,212,243,228,84,236,82,192,131,216,202,196,209,231,167,188,194,227,230,210,216,228,207,199,105,209,109,239,89,221,232,238,203,116,138,234,164,185,224,103,236,230,105,220,86,228,65,188,196,169,224,127,190,236,171,214,181,233,196,183,205,230,86,189,143,193,229,78,216,91,234,64,218,212,183,225,87,228,81,157,168,234,86,221,77,224,123,217,116,213,190,210,202,240,239,227,56,179,231,93,166,238,232,179,204,224,198,234,213,212,203,150,183,194,129,223,69,240,159,223,205,229,203,236,195,219,218,208,215,103,223,225,115,236,123,188,204,219,138,196,131,177,216,119,189,121,239,177,169,196,210,87,178,243,211,126,235,225,188,177,211,176,227,146,204,120,225,229,91,235,230,108,206,179,234,86,234,225,88,238,210,55,158,202,84,203,235,138,206,208,192,203,75,222,233,223,189,198,196,233,115,228,113,216,217,240,184,165,237,207,238,214,209,190,153,228,163,232,39,231,65,157,219,129,233,171,224,101,229,82,190,186,120,162,238,237,235,84,229,244,233,109,183,197,116,202,141,237,239,227,71,207,194,225,97,210,107,199,238,234,208,83,162,170,217,159,162,132,234,90,230,115,163,223,232,224,71,186,181,162,191,186,226,202,231,198,221,60,224,229,230,88,230,236,23,220,181,212,199,190,230,226,219,113,231,236,210,237,66,241,204,215,165,238,136,234,70,226,94,191,194,184,231,209,116,239,123,183,226,49,210,139,233,96,223,58,226,100,208,116,235,235,192,154,225,208,87,199,115,202,225,97,238,111,216,196,211,231,221,213,231,199,173,229,209,223,142,219,75,234,72,231,133,198,154,231,225,106,227,87,188,207,216,165,224,199,231,70,234,213,105,231,226,235,214,211,217,231,235,88,197,186,234,58,212,202,123,221,202,192,225,230,221,117,209,135,229,213,138,176,225,222,79,193,215,145,235,228,108,218,146,218,231,220,106,181,196,109,213,65,235,73,224,67,201,238,217,249,174,221,221,216,193,163,241,231,114,223,73,224,226,136,187,233,222,104,210,123,243,222,241,107,225,74,211,96,192,233,198,231,232,166,232,71,212,219,228,86,225,99,210,118,226,154,180,225,197,232,228,58,203,179,103,224,87,230,59,183,243,93,217,228,232,195,205,227,225,95,236,149,184,225,148,221,79,233,217,206,230,209,223,227,213,236,208,228,226,93,177,231,186,128,219,227,189,230,79,219,189,169,225,92,212,230,179,232,119,212,230,99,222,80,184,202,211,171,237,239,228,230,193,228,111,207,198,108,199,206,119,203,224,119,169,228,223,88,230,230,99,239,146,230,186,203,233,207,65,194,216,215,97,224,239,239,92,221,243,225,57,232,219,185,230,135,195,194,232,224,178,205,216,233,225,88,189,223,202,76,221,207,197,232,227,230,205,228,74,230,64,210,238,222,228,196,212,93,231,77,172,142,217,121,232,148,243,240,184,212,129,175,230,221,165,226,93,172,205,229,219,62,231,223,223,108,241,211,177,220,127,216,105,238,68,246,242,182,218,69,181,235,213,156,228,220,108,230,89,183,203,100,99,232,187,215,234,229,96,186,159,226,184,222,227,59,201,237,100,230,191,231,234,108,184,184,209,188,232,83,211,221,212,219,244,233,157,230,107,219,114,212,237,228,84,178,231,216,84,207,213,224,133,220,83,237,223,233,234,63,188,205,213,159,234,93,223,63,236,95,200,225,173,238,89,229,102,219,149,215,39,230,136,175,219,211,205,177,231,209,229,153,193,138,225,219,80,221,197,180,162,193,123,185,210,224,68,239,79,208,113,225,205,154,228,206,213,230,76,227,113,208,218,171,120,183,163,204,228,98,230,91,235,210,234,94,236,90,216,82,199,226,102,239,227,69,186,133,226,215,236,236,188,222,200,236,185,236,221,237,209,222,114,216,133,206,212,229,219,208,221,210,168,204,211,83,205,196,207,197,93,171,219,225,237,229,109,95,238,231,68,198,230,199,225,225,228,243,75,219,198,205,226,182,230,225,86,199,220,222,84,226,208,238,216,81,212,226,233,237,233,200,193,193,214,158,235,196,228,222,21,191,194,198,221,68,219,211,210,165,212,211,216,196,213,200,215,89,170,209,158,132,226,82,209,222,212,227,75,228,154,194,230,126,226,234,82,237,213,204,229,169,100,219,181,205,208,227,86,237,191,183,233,45,245,231,214,227,86,209,114,220,111,214,162,225,74,235,227,85,235,126,198,240,157,230,220,240,231,102,208,185,237,68,157,142,201,221,99,219,99,227,220,221,214,227,220,110,233,91,189,184,102,233,230,115,209,92,222,233,48,226,62,215,235,91,129,230,236,182,184,226,73,240,209,211,92,241,187,156,205,192,236,113,227,230,202,212,204,232,148,216,80,217,230,223,81,223,80,220,78,213,98,211,188,219,233,69,136,228,107]

content/my_index/ivf.pid.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a5f3d8a075c4b963e8b4f052d8078ca59677ea0708fbc8b42457684a119099c
+size 1677208

content/my_index/metadata.json

@@ -0,0 +1,70 @@
+{
+  "config":{
+    "query_token_id":"[unused0]",
+    "doc_token_id":"[unused1]",
+    "query_token":"[Q]",
+    "doc_token":"[D]",
+    "ncells":null,
+    "centroid_score_threshold":null,
+    "ndocs":null,
+    "load_index_with_mmap":false,
+    "index_path":null,
+    "index_bsize":32,
+    "nbits":8,
+    "kmeans_niters":20,
+    "resume":false,
+    "similarity":"cosine",
+    "bsize":64,
+    "accumsteps":1,
+    "lr":0.00001,
+    "maxsteps":400000,
+    "save_every":null,
+    "warmup":20000,
+    "warmup_bert":null,
+    "relu":false,
+    "nway":64,
+    "use_ib_negatives":true,
+    "reranker":false,
+    "distillation_alpha":1.0,
+    "ignore_scores":false,
+    "model_name":null,
+    "query_maxlen":32,
+    "attend_to_mask_tokens":false,
+    "interaction":"colbert",
+    "dim":128,
+    "doc_maxlen":256,
+    "mask_punctuation":true,
+    "checkpoint":"colbert-ir/colbertv2.0",
+    "triples":"/future/u/okhattab/root/unit/experiments/2021.10/downstream.distillation.round2.2_score/round2.nway6.cosine.ib/examples.64.json",
+    "collection":[
+      "list with 3485 elements starting with...",
+      [
+        "Hyperspectral images (HSIs) have extensive applications in various fields such as medicine agriculture and industry. Nevertheless acquiring high signal-to-noise ratio HSI poses a challenge due to narrow-band spectral filtering. Consequently the importance of HSI denoising is substantial especially for snapshot hyperspectral imaging technology. While most previous HSI denoising methods are supervised creating supervised training datasets for the diverse scenes hyperspectral cameras and scan parameters is impractical. In this work we present Diff-Unmix a self-supervised denoising method for HSI using diffusion denoising generative models. Specifically Diff-Unmix addresses the challenge of recovering noise-degraded HSI through a fusion of Spectral Unmixing and conditional abundance generation. Firstly it employs a learnable block-based spectral unmixing strategy complemented by a pure transformer-based backbone. Then we introduce a self-supervised generative diffusion network to enhance abundance maps from the spectral unmixing block. This network reconstructs noise-free Unmixing probability distributions effectively mitigating noise-induced degradations within these components. Finally the reconstructed HSI is reconstructed through unmixing reconstruction by blending the diffusion-adjusted abundance map with the spectral endmembers.",
+        "This network reconstructs noise-free Unmixing probability distributions effectively mitigating noise-induced degradations within these components. Finally the reconstructed HSI is reconstructed through unmixing reconstruction by blending the diffusion-adjusted abundance map with the spectral endmembers. Experimental results on both simulated and real-world noisy datasets show that Diff-Unmix achieves state-of-the-art performance.",
+        "The reflective nature of the human eye is an under-appreciated source of information about what the world around us looks like. By imaging the eyes of a moving person we capture multiple views of a scene outside the camera's direct line of sight through the reflections in the eyes. In this paper we reconstruct a radiance field beyond the camera's line of sight using portrait images containing eye reflections. This task is challenging due to 1) the difficulty of accurately estimating eye poses and 2) the entangled appearance of the iris textures and the scene reflections. To address these our method jointly optimizes the cornea poses the radiance field depicting the scene and the observer's eye iris texture. We further present a regularization prior on the iris texture to improve scene reconstruction quality. Through various experiments on synthetic and real-world captures featuring people with varied eye colors and lighting conditions we demonstrate the feasibility of our approach to recover the radiance field using cornea reflections."
+      ]
+    ],
+    "queries":"/future/u/okhattab/data/MSMARCO/queries.train.tsv",
+    "index_name":"/content/my_index",
+    "overwrite":false,
+    "root":".ragatouille/",
+    "experiment":"colbert",
+    "index_root":null,
+    "name":"2024-06/16/21.05.35",
+    "rank":0,
+    "nranks":1,
+    "amp":true,
+    "gpus":1,
+    "avoid_fork_if_possible":false
+  },
+  "num_chunks":1,
+  "num_partitions":8192,
+  "num_embeddings":645932,
+  "avg_doclen":185.3463414634,
+  "RAGatouille":{
+    "index_config":{
+      "index_type":"PLAID",
+      "index_name":"/content/my_index"
+    }
+  }
+}

content/my_index/plan.json

@@ -0,0 +1,64 @@
+{
+    "config": {
+        "query_token_id": "[unused0]",
+        "doc_token_id": "[unused1]",
+        "query_token": "[Q]",
+        "doc_token": "[D]",
+        "ncells": null,
+        "centroid_score_threshold": null,
+        "ndocs": null,
+        "load_index_with_mmap": false,
+        "index_path": null,
+        "index_bsize": 32,
+        "nbits": 8,
+        "kmeans_niters": 20,
+        "resume": false,
+        "similarity": "cosine",
+        "bsize": 64,
+        "accumsteps": 1,
+        "lr": 1e-5,
+        "maxsteps": 400000,
+        "save_every": null,
+        "warmup": 20000,
+        "warmup_bert": null,
+        "relu": false,
+        "nway": 64,
+        "use_ib_negatives": true,
+        "reranker": false,
+        "distillation_alpha": 1.0,
+        "ignore_scores": false,
+        "model_name": null,
+        "query_maxlen": 32,
+        "attend_to_mask_tokens": false,
+        "interaction": "colbert",
+        "dim": 128,
+        "doc_maxlen": 256,
+        "mask_punctuation": true,
+        "checkpoint": "colbert-ir\/colbertv2.0",
+        "triples": "\/future\/u\/okhattab\/root\/unit\/experiments\/2021.10\/downstream.distillation.round2.2_score\/round2.nway6.cosine.ib\/examples.64.json",
+        "collection": [
+            "list with 3485 elements starting with...",
+            [
+                "Hyperspectral images (HSIs) have extensive applications in various fields such as medicine agriculture and industry. Nevertheless acquiring high signal-to-noise ratio HSI poses a challenge due to narrow-band spectral filtering. Consequently the importance of HSI denoising is substantial especially for snapshot hyperspectral imaging technology. While most previous HSI denoising methods are supervised creating supervised training datasets for the diverse scenes hyperspectral cameras and scan parameters is impractical. In this work we present Diff-Unmix a self-supervised denoising method for HSI using diffusion denoising generative models. Specifically Diff-Unmix addresses the challenge of recovering noise-degraded HSI through a fusion of Spectral Unmixing and conditional abundance generation. Firstly it employs a learnable block-based spectral unmixing strategy complemented by a pure transformer-based backbone. Then we introduce a self-supervised generative diffusion network to enhance abundance maps from the spectral unmixing block. This network reconstructs noise-free Unmixing probability distributions effectively mitigating noise-induced degradations within these components. Finally the reconstructed HSI is reconstructed through unmixing reconstruction by blending the diffusion-adjusted abundance map with the spectral endmembers.",
+                "This network reconstructs noise-free Unmixing probability distributions effectively mitigating noise-induced degradations within these components. Finally the reconstructed HSI is reconstructed through unmixing reconstruction by blending the diffusion-adjusted abundance map with the spectral endmembers. Experimental results on both simulated and real-world noisy datasets show that Diff-Unmix achieves state-of-the-art performance.",
+                "The reflective nature of the human eye is an under-appreciated source of information about what the world around us looks like. By imaging the eyes of a moving person we capture multiple views of a scene outside the camera's direct line of sight through the reflections in the eyes. In this paper we reconstruct a radiance field beyond the camera's line of sight using portrait images containing eye reflections. This task is challenging due to 1) the difficulty of accurately estimating eye poses and 2) the entangled appearance of the iris textures and the scene reflections. To address these our method jointly optimizes the cornea poses the radiance field depicting the scene and the observer's eye iris texture. We further present a regularization prior on the iris texture to improve scene reconstruction quality. Through various experiments on synthetic and real-world captures featuring people with varied eye colors and lighting conditions we demonstrate the feasibility of our approach to recover the radiance field using cornea reflections."
+            ]
+        ],
+        "queries": "\/future\/u\/okhattab\/data\/MSMARCO\/queries.train.tsv",
+        "index_name": "\/content\/my_index",
+        "overwrite": false,
+        "root": ".ragatouille\/",
+        "experiment": "colbert",
+        "index_root": null,
+        "name": "2024-06\/16\/21.05.35",
+        "rank": 0,
+        "nranks": 1,
+        "amp": true,
+        "gpus": 1,
+        "avoid_fork_if_possible": false
+    },
+    "num_chunks": 1,
+    "num_partitions": 8192,
+    "num_embeddings_est": 645932.0088195801,
+    "avg_doclen_est": 185.34634399414062
+}