diff --git a/src/training/train_gtauav.py b/src/training/train_gtauav.py
index 4e1b16e..7127022 100644
--- a/src/training/train_gtauav.py
+++ b/src/training/train_gtauav.py
@@ -839,15 +839,20 @@ def train(cfg: TrainConfigGTAUAV) -> None:
                 )
             loss_dict = loss_fn(**loss_kwargs)
 
-            # Enqueue current gallery embeddings (detached).
-            if neg_bank is not None:
-                neg_bank.enqueue(embeddings["gallery"].detach())
-
             # Scale loss by accumulation steps so gradients average correctly.
             raw_loss = float(loss_dict["total"].item())  # save before backward
             total_loss = loss_dict["total"] / accum
             scaler.scale(total_loss).backward()
 
+            # Enqueue current gallery AFTER backward. The queue buffer is aliased
+            # into the autograd graph through `queue_neg` (a view returned by
+            # `NegativeMemoryBank.get_queue`), so modifying it before backward
+            # triggers "variable needed for gradient computation has been modified
+            # by an inplace operation". Enqueueing here is semantically identical
+            # — the next step's queue state is the same either way.
+            if neg_bank is not None:
+                neg_bank.enqueue(embeddings["gallery"].detach())
+
             # Optimizer step only after accumulating `accum` micro-batches.
             is_accum_step = (n_batches + 1) % accum == 0 or (n_batches + 1) == len(train_loader)
             if is_accum_step: