[Fix] Qwen2VL batchsize>1 visual alignment

Author: Terry-Uv

lmms_eval/models/simple/qwen2_vl.py

@@ -188,6 +188,16 @@ def _collate(x):
         # Import utils here if flatten is moved
         import lmms_eval.utils as utils
 
+        def _ensure_list(v):
+            if v is None:
+                return []
+            if isinstance(v, list):
+                # incase [[img]]
+                if len(v) == 1 and isinstance(v[0], list):
+                    return v[0]
+                return v
+            return [v]
+
         pbar = tqdm(total=len(requests), disable=(self.rank != 0), desc="Model Responding")
         # we group requests by their generation_kwargs,
         # so that we don't try to execute e.g. greedy sampling and temp=0.8 sampling
@@ -202,13 +212,10 @@ def _collate(x):
             # TODO: Clarify the behavior of doc_to_visual for documents without visual info.
             # The current logic might incorrectly discard all visuals if one doc lacks them.
             # Ensure flatten is appropriate here based on doc_to_visual's return type.
-            visual_list = [doc_to_visual[0](self.task_dict[task][split][ids]) for ids in doc_id]
-            if None in visual_list:  # This check might need refinement
-                # If a mix of visual/non-visual is possible, this needs careful handling
-                # Currently sets all visuals to empty if any doc returns None
-                visual_list = []
-            else:
-                visual_list = self.flatten(visual_list)  # Assumes doc_to_visual returns list of lists
+            visuals_per_doc = []
+            for fn, ids in zip(doc_to_visual, doc_id):
+                v = fn(self.task_dict[task][split][ids])
+                visuals_per_doc.append(_ensure_list(v))
 
             gen_kwargs = all_gen_kwargs[0] if all_gen_kwargs else {}
 
@@ -249,7 +256,7 @@ def _collate(x):
                 # Needs careful review based on doc_to_visual output structure
                 # For simplicity, assuming visual_list contains all visuals for the batch for now
                 # A more robust approach might map visuals back to their original context index.
-                relevant_visuals = visual_list  # Placeholder: needs logic to get visuals for context 'i'
+                relevant_visuals = visuals_per_doc[i]  # Placeholder: needs logic to get visuals for context 'i'
 
                 for visual in relevant_visuals:
                     if isinstance(visual, str) and visual.endswith((".mp4", ".avi", ".mov")):  # Video file